14.12.2012 Views

Stanton PhD Thesis final_docx - Atrium - University of Guelph

Stanton PhD Thesis final_docx - Atrium - University of Guelph

Stanton PhD Thesis final_docx - Atrium - University of Guelph

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

AN EVALUATION OF THE IMPACT OF MANAGEMENT PRACTICE ON THE<br />

HEALTH AND WELFARE OF DAIRY HEIFER CALVES<br />

A thesis<br />

Presented to<br />

The Faculty <strong>of</strong> Graduate Studies<br />

<strong>of</strong><br />

The <strong>University</strong> <strong>of</strong> <strong>Guelph</strong><br />

by<br />

AMY STANTON<br />

In partial fulfillment <strong>of</strong> requirements<br />

for the degree <strong>of</strong><br />

Doctor <strong>of</strong> Philosophy<br />

July, 2011<br />

©Amy <strong>Stanton</strong>, 2011


ABSTRACT:<br />

AN EVALUATION OF THE IMPACT OF MANAGEMENT PRACTICE ON THE<br />

HEALTH AND WELFARE OF DAIRY HEIFER CALVES<br />

Amy <strong>Stanton</strong> Co-Advisors:<br />

<strong>University</strong> <strong>of</strong> <strong>Guelph</strong>, 2011 David Kelton & Suzanne T Millman<br />

The objectives <strong>of</strong> this thesis were to investigate 1) the use <strong>of</strong> behavior and activity<br />

monitoring for the identification <strong>of</strong> heifers at risk <strong>of</strong> disease, 2) the use <strong>of</strong> group level<br />

management practices to reduce the risk <strong>of</strong> disease, and 3) the identification <strong>of</strong> long-term<br />

impacts <strong>of</strong> bovine respiratory disease complex (BRD).<br />

For objective 1, lying posture, a decreased willingness to approach an observer and high<br />

lethargy scores were associated with diarrhea in calves under 2 weeks <strong>of</strong> age and a high lethargy<br />

score in 4-6 week old calves was associated with decreased average daily gain (ADG) in the first<br />

8 weeks <strong>of</strong> life (n = 744). In weaned calves (n = 74) increased activity (increase in steps and<br />

decrease in lying), standing at the bunk not eating, and lying far from other calves in the first 3<br />

days post-weaning were associated with decreased post-weaning weight gain.<br />

For objective 2, separating social mixing from movement to a novel environment, and<br />

administering prophylactic antibiotics to calves at high risk <strong>of</strong> disease, were investigated. Both<br />

mixing and movement to a novel environment increased activity levels in newly weaned dairy<br />

calves (n = 64). When calves were mixed prior to movement to a novel environment they had a<br />

smaller increase in activity compared to calves that were simultaneously mixed and moved. No


differences in weight gain or calf starter intake were observed. Administration <strong>of</strong> a prophylactic<br />

antibiotic, tulathromycin, to 3 day old calves upon arrival at a heifer raising facility (n = 788) and<br />

8 week old calves at first movement to group housing (n = 1,392 ) was found to reduce diarrhea<br />

and otitis media, and BRD, respectively.<br />

Objective 3 was addressed by monitoring calves that received tulathromycin at 8 weeks<br />

<strong>of</strong> age to determine the long-term impacts <strong>of</strong> BRD. Bovine Respiratory Disease complex was<br />

associated with decreased growth to 9 months <strong>of</strong> age, decreased survival to first calving,<br />

increased risk <strong>of</strong> dystocia and a greater age at first calving.


Table <strong>of</strong> Contents<br />

CHAPTER 1: LITERATURE REVIEW 1<br />

1.1 Introduction 1<br />

1.2 The use <strong>of</strong> behavior for early disease detection 3<br />

1.3 Review <strong>of</strong> current management practices to reduce disease 6<br />

1.4 Stress and the dairy calf 8<br />

1.4.1 Background 8<br />

1.4.2 Nutritional Stressors 10<br />

1.4.3 Social Stressors 13<br />

1.4.4 Transport Stress 18<br />

1.5 Control <strong>of</strong> Disease Spread 19<br />

1.6 Long-term impacts <strong>of</strong> Bovine Respiratory Disease Complex 21<br />

1.7 Conclusion and thesis objectives 24<br />

1.8 References 25<br />

CHAPTER 2: PREVALENCE OF SICKNESS BEHAVIOR IN NEONATAL DAIRY CALVES AT A<br />

COMMERCIAL HEIFER FACILITY AND ASSOCIATIONS WITH AVERAGE DAILY GAIN AND<br />

ILLNESS 33<br />

2.1 Introduction 33<br />

2.2 Materials & Methods 35<br />

2.2.1 Animals and Housing 35<br />

2.2.2 Behavioral Measurements 36<br />

2.2.3 Disease Recordings and Weight Gain 39<br />

2.2.4 Statistical Analysis 40<br />

2.3 Results 42<br />

2.3.1 Changes in response to behavior measures over time and associations with<br />

Average Daily Gain 42<br />

2.3.2 Behaviors associated with neonatal calf diarrhea complex 43<br />

2.4 Discussion 44<br />

2.5 References 52<br />

CHAPTER 3: ASSOCIATIONS AMONG ACTIVITY PATTERNS AND WEIGHT GAIN IN NEWLY<br />

WEANED DAIRY CALVES 63<br />

3.1 Introduction 63<br />

3.2 Materials and Methods 65<br />

3.2.1 Behavior Measurements 67<br />

iv


3.2.2 Experimental Study Design 68<br />

3.2.3. Statistical Analysis 68<br />

3.3 Results 70<br />

3.3.1 Activity level in response to weaning and relocation 71<br />

3.3.2 Bunk behavior 71<br />

3.3.3 Resting Behavior 73<br />

3.4 Discussion 73<br />

3.5 References 79<br />

CHAPTER 4: EVALUATION OF A GRADUATED WEANING PRACTICE ON BEHAVIOR AND<br />

PERFORMANCE OF DAIRY CALVES 87<br />

4.1 Introduction 87<br />

4.2 Materials and Methods 89<br />

4.2.1 Pre-enrolment management 89<br />

4.2.2. Housing 89<br />

4.2.3. Study Treatment 90<br />

4.2.4. Blood Sampling 91<br />

4.2.5 Data Collection 92<br />

4.2.6 Statistical Analysis 92<br />

4.3 Results 94<br />

4.3 Discussion 96<br />

4.4 References 98<br />

CHAPTER 5: THE EFFECT OF TREATMENT WITH LONG-ACTING ANTIBIOTIC UPON<br />

ARRIVAL AT A HEIFER RAISING FACILITY ON DISEASE AND GROWTH IN COMMERCIAL<br />

DAIRY HEIFERS 103<br />

5.1 Abstract 103<br />

5.2 Introduction 105<br />

5.3 Materials and Methods 107<br />

5.3.1 Heifer Management and Housing 107<br />

5.3.2 Study Treatments 108<br />

5.3.3 Outcome Measures 109<br />

5.3.4 Sampling for Mycoplasma testing 109<br />

5.3.5 Statistical Analysis 110<br />

5.4 Results 112<br />

5.4.1 Health Events in the Study Population 112<br />

5.4.2 Morbidity 112<br />

5.4.3 Growth 114<br />

5.4.4 Height at movement to group housing 115<br />

v


5.4.4 Testing for Mycoplasma 116<br />

5.4.4 Mortality 117<br />

5.5. Discussion 117<br />

5.6 References 124<br />

CHAPTER 6: THE EFFECT OF TREATMENT WITH LONG-ACTING ANTIBIOTIC AT<br />

POSTWEANING MOVEMENT ON RESPIRATORY DISEASE AND ON GROWTH IN<br />

COMMERCIAL DAIRY CALVES 133<br />

6.1 Abstract 134<br />

6.2 Introduction 135<br />

6.3 Materials and Methods 137<br />

6.3.1. Heifer Management and Housing 137<br />

6.3.2 Study Treatments 138<br />

6.3.3 Outcome Measures 140<br />

6.3.4 Statistical Analysis 141<br />

6.4 RESULTS 142<br />

6.4.1 Incidence <strong>of</strong> BRD in the 60 d Post-Movement 142<br />

6.4.2 Growth in Postweaned Dairy Calves 143<br />

6.4.3 Death Loss 144<br />

6.5 Discussion 145<br />

6.6. Acknowledgments 149<br />

6.7 References 150<br />

CHAPTER 7: THE IMPACT OF BOVINE RESPIRATORY DISEASE AND A PREVENTATIVE<br />

ANTIBIOTIC TREATMENT ON GROWTH, SURVIVAL, AGE AT FIRST CALVING, AND<br />

PRODUCTIVITY HEIFERS 158<br />

7. 1 Abstract 158<br />

7.2 Introduction 160<br />

7.3 Material and Methods 162<br />

7.3.1 Heifer Management and Housing 162<br />

7.3.2 Outcome Measures 163<br />

7.3.3 Statistical Analysis 165<br />

7.3.4 Exclusion Criteria 167<br />

7.4 Results 168<br />

7.4.1 Average Daily Gain 168<br />

7.4.2 Body weight 169<br />

7.4.3 Height 169<br />

7.4.4 Survival to First Calving 169<br />

vi


7.4.5 Culling and mortality prior to first calving 170<br />

7.4.6 Age at First Calving 171<br />

7.4.7 Dystocia 171<br />

7.4.8 Milk Production 172<br />

7.4.9 Survival to 120 days in milk 173<br />

7.5 Discussion 173<br />

7.6 Acknowledgements 179<br />

7.7 References 180<br />

CHAPTER 8: CONCLUSIONS AND FUTURE RESEARCH 198<br />

8.1 General Discussion 198<br />

8.2 Future Research 203<br />

8.3 References 207<br />

vii


List <strong>of</strong> Tables<br />

Table 2.1: Ethogram <strong>of</strong> behaviors recognized and recorded during behavioral scoring .............. 55<br />

Table 2.2: List <strong>of</strong> exclusions used to select healthy controls for comparison to calves with active<br />

diarrhea ......................................................................................................................................... 56<br />

Table 2.3: Prevalence <strong>of</strong> behavioral responses among 692 milk-fed individually housed heifer<br />

calves at a commercial heifer raising facility at three observation periods (average age 10, 24 and<br />

38 days, respectively) with the odds <strong>of</strong> the response occurring at period 2 and 3, relative to<br />

period 1, controlling for calf as a random effect. .......................................................................... 57<br />

Table 2.5: Proportional responses <strong>of</strong> diarrhea and control calves to the observer during vigilance<br />

testing ............................................................................................................................................ 59<br />

Table 3.1: Descriptive list <strong>of</strong> observed behaviours ...................................................................... 80<br />

Table 3.2: Associations between activity and Day 7 weight <strong>of</strong> calves in the week post-weaning.<br />

All models included Day 0 weight and nursery Locations in the model as a fixed effect and<br />

group nested within group size as a random effect. All behaviours (steps, lying duration, daily<br />

lying bouts) were assessed independently <strong>of</strong> each other. ............................................................. 81<br />

Table 4.1: Number <strong>of</strong> calves identified with bovine respiratory disease complex (BRD) by<br />

treatment group and study time period. ........................................................................................ 99<br />

Table 5.1: Disease definitions used by farm staff ...................................................................... 127<br />

Table 5.2: Incidence and median age at first treatment for all health events for calves at a<br />

commercial heifer raising facility during the first 8 weeks <strong>of</strong> life. ............................................. 128<br />

Table 5.3: Incidence <strong>of</strong> health events for calves sampled for Mycoplasma at 0, 2 and 4 weeks <strong>of</strong><br />

age. .............................................................................................................................................. 129<br />

Table 5.4: Breakdown <strong>of</strong> the number <strong>of</strong> positive M. Bovis samples collected from calves at<br />

repeated intervals (0, 2, and 4) by source farm and strain type .................................................. 130<br />

Table 5.5: Incidence <strong>of</strong> health conditions as reported in the literature for calves under 6 months<br />

<strong>of</strong> age ........................................................................................................................................... 131<br />

Table 6.1: Disease variables considered in models for weight, height, and risk <strong>of</strong> bovine<br />

respiratory disease (BRD) ........................................................................................................... 152<br />

Table 6.2: The probability <strong>of</strong> respiratory disease in the 6 wk after movement in 1,392 weaned<br />

dairy calves, randomly assigned to treatment with tulathromycin (TUL) or oxytetracycline (TET)<br />

at the time <strong>of</strong> movement to group housing 1 ................................................................................ 153<br />

Table 6.3: Least squares means ( ± SEM) <strong>of</strong> the weight (kg) <strong>of</strong> commercial heifers recorded<br />

upon exit from the weaning barn controlling for age at <strong>final</strong> measurement and weight at<br />

enrollment with random effects for source farm and weekly enrollment cohort ........................ 154<br />

Table 6.4: Least squares means ( ± SEM) <strong>of</strong> the height (cm) <strong>of</strong> commercial heifers upon exit<br />

from the weaning barn controlling for age at <strong>final</strong> measurement and height at enrollment with<br />

random effects for source farm and weekly enrollment cohort .................................................. 155<br />

viii


Table 6.5: Reasons for death loss and culling in the 60 d following movement to group housing<br />

..................................................................................................................................................... 156<br />

Table 7.1: Calving Ease Scoring System recorded in farm DairyComp records and reported to<br />

DHIA........................................................................................................................................... 182<br />

Table 7.2: Weight and height <strong>of</strong> 1,392 calves upon entry in the 6 rearing barns at the commercial<br />

heifer raising facility (mean ± S.D.) ........................................................................................... 183<br />

Table 7.3: Least-square means (± SE) <strong>of</strong> the average daily gain (kg/day) <strong>of</strong> 1,392 heifers<br />

recorded during housing in 4 barns and between enrolment and exit from Grower Barn 2,<br />

controlling for age at exit from each barn, birth weight, pre-enrolment average daily gain, preenrolment<br />

bovine respiratory disease and season with random effects for source farm and weekly<br />

enrolment cohort. ........................................................................................................................ 184<br />

Table 7.4: Differences <strong>of</strong> least squares means (± SE) in the wither height (cm) between calves<br />

administered tulathromycin or oxytetracycline at enrolment at a commercial heifer raising<br />

facility. The linear mixed model controlled for age at <strong>final</strong> measurement, bovine respiratory<br />

disease prior to trial enrolment (pBRD), height at birth, season, and random effects for source<br />

farm and weekly enrollment cohort ............................................................................................ 185<br />

Table 7.5: Proportion <strong>of</strong> calves surviving to first calving by source farm and bovine respiratory<br />

disease status ............................................................................................................................... 186<br />

Table 7.6: Producers’ primary reason for the 7% (93/1293) <strong>of</strong> heifers that died between<br />

enrolment (8weeks <strong>of</strong> age) and calving. ..................................................................................... 187<br />

Table 7.7: Producers primary reason for the 12% (162/1,392) <strong>of</strong> heifers culled from herd<br />

between enrolment (8weeks <strong>of</strong> age) and calving. ....................................................................... 188<br />

Table 7.8: Distribution <strong>of</strong> calving ease score recorded by source farms in 1066 heifers .......... 189<br />

Table 7.9: Effect <strong>of</strong> BRD60 on calving ease score using multiple cut-points as evaluated in a<br />

logistic model with random effects for source farm and enrolment cohort. ............................... 190<br />

Table 7.10: The number <strong>of</strong> cows with and without bovine respiratory disease in the 60 days<br />

following enrollment (BRD60) from each source farm in the analysis <strong>of</strong> first milk test, the leastsquares<br />

mean (± standard error <strong>of</strong> the mean) age at first milk test and the effect <strong>of</strong> BRD60 on<br />

first milk test date........................................................................................................................ 191<br />

Table 7.11: Distribution <strong>of</strong> heifers that survived to 120 days post-calving recorded by calving<br />

ease score and the probability <strong>of</strong> survival relative to a calving ease score <strong>of</strong> 1 ......................... 192<br />

ix


List <strong>of</strong> Figures<br />

Figure 2.1: Timeline describing trial milestones and behavioral observations, with average ages<br />

(mean ± s.d.) at observation and measurements ........................................................................... 60<br />

Figure 2.2: Method for behavioral scoring <strong>of</strong> the calves, conducted every other week .............. 61<br />

Figure 2.3: Proportion <strong>of</strong> calves in each lethargy response category for the active diarrhea (n =<br />

229) and healthy controls (n = 47) ................................................................................................ 62<br />

Figure 3.1: The average number <strong>of</strong> steps per day (mean ± standard deviation ) by weight gain<br />

category in the first week post-weaning for calves (n = 70) following weaning and relocation to a<br />

novel environment. ....................................................................................................................... 82<br />

Figure 3.2: The total lying duration per day (hours) (mean ± standard deviation ) for calves (n =<br />

70) following weaning and relocation to a novel environment, and total lying duration per day by<br />

weight gain category in the first week post-enrolment ................................................................. 83<br />

Figure 3.3: The average number <strong>of</strong> lying bouts (mean ± standard deviation ) per day for calves<br />

(n = 70) following weaning and relocation to a novel environment, and number <strong>of</strong> lying bouts<br />

per day by weight gain category in the first week post-enrolment ............................................... 84<br />

Figure 3.4: LSMEANS (±S.E.) <strong>of</strong> post-weaning weight gain for total lying duration by nursery<br />

location interaction (P = 0.01) ...................................................................................................... 85<br />

Figure 3.5: LSMEANS (±S.E.) <strong>of</strong> post-weaning weight gain for time at the bunk (% <strong>of</strong> total<br />

observations) by nursery location interaction (P = 0.03), controlling for Day 0 weight, antibiotic<br />

treatment, and pre-weaning calf starter intake, with random effects for group nested within group<br />

size. ............................................................................................................................................... 86<br />

Figure 4.1: Steps per day (Mean and 95% Confidence Interval ) for each <strong>of</strong> the time periods,<br />

controlling for breed and age as fixed effects, replicate as a random effect and calf nested within<br />

replicate and treatment as a repeated measure. ........................................................................... 100<br />

Figure 4.2: LSMeans ( ± S.E.) <strong>of</strong> total daily lying time (hours), controlling for breed and age<br />

with a random effect for replicate and calf nested within replicate and treatment. .................... 101<br />

Figure 5.1: Images taken from farm prior to the start <strong>of</strong> the trial and used as guidelines to<br />

identify ear droop. Shown: from left to right: Unilateral ear droop and ear droop with abnormal<br />

head position ............................................................................................................................... 132<br />

Figure 6.1: Timeline <strong>of</strong> trial events ............................................................................................ 157<br />

Figure 7.1: Timeline <strong>of</strong> trial events ............................................................................................ 193<br />

Figure 7.2: Difference in least squares means (± SE ) <strong>of</strong> the weight (kg) <strong>of</strong> heifers with BRD60 1<br />

relative to calves without BRD60 at 5 time points at a commercial heifer raising facility,<br />

controlling for birth weight, age at measurement and season with random effects for source farm<br />

and weekly enrolment cohort. ..................................................................................................... 194<br />

Figure 7.3: Differences <strong>of</strong> least squares means (± SE ) <strong>of</strong> the weight (kg) <strong>of</strong> heifers with no prior<br />

history <strong>of</strong> bovine respiratory disease that received tulathromycin at study enrolment relative to<br />

heifers that received oxytetracycline at study enrolment at entry at 5 time points at a commercial<br />

heifer raising facility. The linear mixed models controlled for birth weight, age at measurement<br />

x


and season as fixed effects and for source farm and weekly enrolment cohort as random effects.<br />

..................................................................................................................................................... 195<br />

Figure 7.4: Time to event analysis for age at first calving stratified by BRD60 status with culled<br />

heifers censored (P = 0.02). ........................................................................................................ 196<br />

xi


CHAPTER 1: LITERATURE REVIEW<br />

1.1 INTRODUCTION<br />

Bovine respiratory disease complex (BRD) and neonatal calf diarrhea complex are the<br />

two most commonly diagnosed diseases in pre-weaned and weaned dairy calves, and account for<br />

between 60-80% <strong>of</strong> dairy heifer mortality in the United States (USDA, 2010b). These diseases<br />

are very important due to the impact that they have on animal welfare and productivity.<br />

Animal welfare issues can be categorized into three main ethical concerns (Fraser et al.,<br />

1997). The first category is functional, which are concerns focused on deviations from normal<br />

physiologic function, such as poor health or growth. The second category stems from a feelings<br />

based approach to animal welfare, which are concerns focused on avoiding prolonged or intense<br />

pain, discomfort or other negative emotional states, as well as allowing for positive emotional<br />

states. The <strong>final</strong> ethical category is based on the natural view, which focuses on the natural<br />

behaviors <strong>of</strong> animals based on their evolutionary adaptations and capabilities. Although<br />

described separately, in practice many <strong>of</strong> the current animal welfare concerns negatively impact<br />

animals under all <strong>of</strong> the ethical areas. An example <strong>of</strong> this would be bovine respiratory disease<br />

(BRD), which decreases the health and growth <strong>of</strong> animals. BRD also is likely to be associated<br />

with shortness <strong>of</strong> breath, or ataxia, which is considered to be a negative affective state in many<br />

species (Mellor & Stafford, 2004). By considering all three ethical concerns in conjunction with<br />

the potential economic impact for the producer, management systems can be improved for the<br />

benefit <strong>of</strong> both producers and animals.<br />

The failure <strong>of</strong> the dairy industry to address animal welfare concerns can lead to<br />

legislative changes, which can force change. An example <strong>of</strong> this is the European Union directive<br />

1


, which requires that “all calves be group housed by 8 weeks <strong>of</strong> age and prior to this stalls must<br />

not have solid walls, but perforated walls , which allow the calves to have direct visual and<br />

tactile contact” (91/629/EEC). In 2006, the state <strong>of</strong> Arizona passed Proposition 204 (The<br />

Humane Treatment <strong>of</strong> Farm Animal Act) , which will require sows and veal calves to be able to<br />

“lie down and fully extending its limbs or turn around freely” and will require compliance by<br />

December 31 st , 2012. California’s Proposition 2 (Treatment <strong>of</strong> Farm Animals Act) passed in<br />

2008, which requires minimum space allowances for calves based on their ability to “turn around<br />

freely, lie down, stand up, and fully extend their limbs” will require compliance by January 1 st<br />

2015.<br />

Voluntary industry changes have been adopted by some parts <strong>of</strong> the calf rearing<br />

agriculture sectors in order to provide guidelines for ethical practices and to avoid legislative<br />

changes made by the public, who are less likely to be familiar with the background <strong>of</strong> many<br />

practices. Examples <strong>of</strong> this include the American Veal Association’s decision to require group<br />

housing by December 31, 2017 (American Veal Association, 2007) and the 2009 Canadian<br />

Codes <strong>of</strong> Practice for the Care and Handling <strong>of</strong> Dairy Cattle , which require that calves be able to<br />

“easily stand up, lie down, turn around and adopt normal resting postures, and have visual<br />

contact with other calves” (National Farm Animal Council, 2009).<br />

Although no legislation has specifically targeted disease rates in dairy calves, this<br />

represents an important area <strong>of</strong> concern for the dairy industry. The National Animal Health<br />

Monitoring System (NAHMS) reported that the average mortality rate are 7.8% and 1.8% in pre-<br />

weaned and weaned dairy calves, respectively (USDA, 2009). In comparison, the NAHMS<br />

survey <strong>of</strong> beef cow-calf operations found that pre-weaned calf mortality was 3.2% with 56% <strong>of</strong><br />

2


operations reporting 0% mortality (USDA, 2010a). This high mortality rate represents both a<br />

welfare issue and an economic concern for producers.<br />

The objectives <strong>of</strong> this paper are to review the current status <strong>of</strong> the literature relating to the<br />

use <strong>of</strong> behavior and automated measures for the identification <strong>of</strong> animals at risk <strong>of</strong> disease, the<br />

use <strong>of</strong> group level management to reduce the risk <strong>of</strong> neonatal calf diseases, and the identification<br />

<strong>of</strong> long-term impacts <strong>of</strong> respiratory diseases. These factors will be reviewed in the context <strong>of</strong> the<br />

current management systems in North America and the associated animal welfare implications.<br />

1.2 THE USE OF BEHAVIOR FOR EARLY DISEASE DETECTION<br />

Disease detection in animals is frequently based on a coordinated set <strong>of</strong> responses to<br />

infection collectively termed sickness behavior (Hart, 1988). Sickness behavior includes the<br />

following behaviours and symptoms: malaise, depression, fever, increased sleep, anorexia,<br />

adipsia, decreased exploratory behavior and decreased socializing. Although initially categorized<br />

as a maladaptive response to inflammation and infection, Hart (1988) proposed that these<br />

behavioral changes are actually an adaptive response that improves the host’s ability to survive<br />

colonization by viruses and bacteria. These responses have further been proposed to be based on<br />

motivational changes, which have very important implications for the application <strong>of</strong> behavioral<br />

means <strong>of</strong> disease detection in the field (Aubert, 1999). This motivational organization <strong>of</strong> sickness<br />

behaviour was determined by a series <strong>of</strong> experiments examining the impact <strong>of</strong> different<br />

physiological and environmental triggers on the responses <strong>of</strong> rats and mice to infection. Key<br />

findings from this research were that maternal behavior, nest building and pup retrieval, were<br />

performed in dams challenged with lipopolysaccharide (LPS), which triggers a febrile response,<br />

in cooler environmental temperatures. However, at thermoneutral temperatures, the maternal<br />

3


ehaviour <strong>of</strong> dams challenged with LPS was altered and nest building was not performed. This<br />

finding supported the hypothesis that these behaviours were driven by motivation changes rather<br />

than reflexes since nest building was not performed in warm environments where the lack <strong>of</strong> a<br />

nest would have minimal impact on pup survival. As such, motivational changes need to be<br />

considered when using sickness behavior to identify diseased animals.<br />

Additional factors that have been shown to alter the expression <strong>of</strong> sickness behavior are<br />

reproductive status, social status and gender. The role <strong>of</strong> reproductive status in the expression <strong>of</strong><br />

sickness behavior has been shown to interact in free-living song sparrows. Male song sparrows, ,<br />

which are territorial year round, showed decreased territorial aggression during the winter (non-<br />

breeding) months in response to a LPS challenge but did not show alterations in aggression<br />

during the spring (breeding) months (Owen-Ashley and Wingfield, 2006). The effect <strong>of</strong> gender<br />

on sickness behaviour has also been examined in rats where it was determined that injections<br />

with LPS altered the sexual receptiveness, attractiveness, and drive <strong>of</strong> female rats but did not<br />

alter these characteristics in male rats (as reviewed by Avitsur and Yirmiya, 1999). This led to<br />

the hypothesis that either interleukin-1 or LPS induced the suppression <strong>of</strong> ovarian hormone<br />

secretion, which may have important implications for dairy cattle given the negative impact on<br />

reproduction.<br />

In rodent models, sickness behaviour suppresses social exploration. In laboratory rats and<br />

mice, illness reduced the amount <strong>of</strong> time animals spent socially exploring unknown juveniles<br />

(Bluthe´ et al., 1992; Bluthé et al., 1994). Social rank also appears to play a role in the display <strong>of</strong><br />

sickness behaviour. When a LPS injection was administered to either a dominant or subordinate<br />

animal in a pair <strong>of</strong> mice, only the dominant mice displayed a reduction in social and agonistic<br />

4


ehavior (Cohn and de Sá-Rocha, 2006). This led to the hypothesis that social status may alter<br />

the expression <strong>of</strong> sickness behavior, especially in confined situations.<br />

Anorexia, adipsia and decreased exploratory behavior associated with classic sickness<br />

behavior. Where treatment <strong>of</strong> illness is not available these behaviours are adaptive since they<br />

promote energy conservation which is needed for the febrile response (Hart, 1988). However, in<br />

modern animal agriculture, where treatment is occurring, this response is <strong>of</strong>ten considered to be<br />

detrimental to growth and recovery rather than adaptive. These behavioural changes may be<br />

especially problematic for calves post-weaning. When individually housed calves are moved to<br />

group housing they need to locate and consume feed and water in a novel environment. If their<br />

motivation to locate and consume feed and water is reduced due to sickness associated<br />

behavioral changes they will likely be unable to maintain adequate nutrient intake. Additionally,<br />

water and grain are necessary for continued rumen development (Kehoe et al., 2007) so animals<br />

that are anorexic and consuming less water will be slowing the necessary development <strong>of</strong> their<br />

rumen.<br />

It has also been hypothesized that the evolutionary role <strong>of</strong> a species will impact the<br />

expression <strong>of</strong> sickness behavior. Cattle, as prey animals, have been identified as ‘stoic’ in<br />

response to pain and disease (Weary et al., 2009). The expression <strong>of</strong> obvious illness in prey<br />

animals is likely to draw predatory attention and increase the risk <strong>of</strong> the sick animal being<br />

targeted. This hypothesis has not been formally tested, but it does raise the important issue<br />

regarding the role <strong>of</strong> species on the expression <strong>of</strong> sickness behavior and indicates the need for<br />

research into species specific sickness behaviors for the detection <strong>of</strong> disease.<br />

5


In cattle, early disease detection has been identified as a key component <strong>of</strong> successful<br />

disease management. Automated monitoring <strong>of</strong> health and behavior has been studied for the<br />

improvement <strong>of</strong> disease detection. These automated monitoring systems have been designed in<br />

part to overcome the difficulties in identifying sick cattle due to their stoic nature and the<br />

decreasing availability <strong>of</strong> skilled farm labour. Examples <strong>of</strong> automated disease detection include<br />

the use <strong>of</strong> infrared thermography to detect increased body temperature (Schaefer et al., 2007) and<br />

monitoring <strong>of</strong> eating and drinking (Buhman et al., 2000; Schaefer et al., 2007) in beef calves for<br />

the early detection <strong>of</strong> BRD. In dairy cattle, changes in feeding, drinking, lying and social<br />

behavior have been used as early behavioral indicators in cows with a variety <strong>of</strong> transition<br />

diseases including metritis (Huzzey et al., 2007), dystocia (Proudfoot et al., 2009), and<br />

subclinical ketosis (Huzzey et al., 2007; Goldhawk et al., 2009; Proudfoot et al., 2009). In dairy<br />

calves, automated milk feeder technology has been used to identify sick calves by changes in<br />

feeding behavior (Svensson and Jensen, 2007; Borderas et al., 2008; Borderas et al., 2009).<br />

These findings have been very useful and have the potential to greatly improve disease detection<br />

among group housed, milk-fed dairy calves. However, in the United States 75% <strong>of</strong> operations<br />

house milk-fed dairy calves individually (USDA, 2007); automated feeders are expensive and as<br />

such studies that depend on them have limited application for many producers. In addition, these<br />

studies do not apply to newly weaned calves , which are also at risk <strong>of</strong> clinical disease.<br />

1.3 REVIEW OF CURRENT MANAGEMENT PRACTICES TO REDUCE DISEASE<br />

In the United States, 97% <strong>of</strong> dairy calves are separated from their mother within 24 hours<br />

<strong>of</strong> birth (USDA, 2009). After birth, 75% <strong>of</strong> operations house calves individually, <strong>of</strong>ten with<br />

minimal physical contact between animals until calves have been weaned from milk-based diets.<br />

6


These two practices have been recommended to reduce the spread <strong>of</strong> disease between adult cattle<br />

and neonates.<br />

The two most common diseases in dairy calves are neonatal calf diarrhea complex and<br />

bovine respiratory disease complex. Neonatal calf diarrhea complex is a broad category <strong>of</strong><br />

diseases , which occurs in approximately 24% <strong>of</strong> dairy calves in the United States (USDA,<br />

2010b). The main diarrheal agents in dairy calves are enterotoxigenic Escherichia coli K99,<br />

Salmonella spp, bovine rotavirus, bovine coronavirus, and Cryptosporidum parvum, which may<br />

act independently or concurrently (Snodgrass et al., 1986; de la Fuente et al., 1999). These<br />

pathogens can cause temporary changes and damage to the intestinal tract through destruction<br />

and sloughing <strong>of</strong> enterocytes, villous atrophy, and inflammation <strong>of</strong> the submucosa, causing<br />

inflammation and sensitization <strong>of</strong> the visceral afferent nerves. In people, these changes to the<br />

intestinal tract are associated with abdominal pain and discomfort (Al-Chaer and Traub, 2002).<br />

The primary treatment for diarrhea is supportive fluid replacement therapy. However,<br />

meloxicam (Metacam, Boehringer Ingelheim (NZ) Ltd), a non-steroidal anti-inflammatory drug,<br />

has been shown to be effective in reducing recovery time from diarrhea when provided in<br />

conjunction with fluid replacement therapy (Todd, 2007). This is a novel approach to treating<br />

calf hood disease, since it is aimed at reducing the discomfort and inflammation associated with<br />

diarrhea rather than a treatment that targets disease causing organisms. This represents an<br />

important change in disease management and animal welfare since it addresses not only the<br />

physiologic signs, but also the discomfort likely experienced by the calves during clinical<br />

disease. The discomfort felt by calves during a diarrhea outbreak is likely to reduce activity<br />

which would promote convalescence. However, if animals are being successfully treated, the<br />

7


increased rest and decreased appetite associated with intestinal discomfort would be unnecessary<br />

since the population <strong>of</strong> disease causing organisms would already be reduced.<br />

The long term consequences <strong>of</strong> diarrhea include increased risk <strong>of</strong> being sold prior to<br />

calving, increased age at first calving (Waltner-Toews et al., 1986), and lower first lactation<br />

energy corrected 305 d milk production (Svensson and Hultgren, 2008). The combination <strong>of</strong><br />

production and welfare impacts makes it critical that management practices are instituted to<br />

decrease disease spread and the susceptibility <strong>of</strong> calves to disease, and identify and manage<br />

affected animals to reduce the negative consequences <strong>of</strong> this disease.<br />

To reduce the incidence <strong>of</strong> disease, three approaches can be used: improve the resistance<br />

<strong>of</strong> the animal to the disease causing agents, reduce or eliminate the disease causing agent, or alter<br />

the environment so that the animal cannot come in contact with disease causing agents.<br />

Examples <strong>of</strong> this theory currently practiced in the dairy industry include reducing common<br />

stressors , which may inhibit the immune response <strong>of</strong> calves making them more likely to<br />

succumb to disease, reducing the spread <strong>of</strong> disease causing organisms by vaccination and early<br />

treatment, and promoting housing and sanitary practices that limits the exposure <strong>of</strong> calves to<br />

disease.<br />

1.4 STRESS AND THE DAIRY CALF<br />

1.4.1 Background<br />

Stress can be classified into two categories, acute and chronic stress. Acute stress is a<br />

short-term stress that is contained within minutes or hours. A chronic stress persists for longer<br />

periods <strong>of</strong> times, anywhere from several hours per day to months at a time (Dhabhar, 2002).<br />

8


When a stressful stimulus occurs, a stress response is initiated via an increase in the cyclic<br />

pulsations <strong>of</strong> corticotrophin releasing hormone (CRH) and arginine-vasopressin neurons <strong>of</strong> the<br />

paraventricular nuclei <strong>of</strong> the hypothalamus. This activation <strong>of</strong> the Hypothalamic–Pituitary–<br />

Adrenal (HPA) axis results in increased adrenocorticotropic hormone (ACTH) and cortisol<br />

secretory episodes. ACTH is the key regulator <strong>of</strong> glucocorticoid secretions (Tsigos and<br />

Chrousos, 2002). Activation <strong>of</strong> the sympathetic nervous system occurs simultaneously to HPA<br />

activation and redirects blood flow from the interior organs <strong>of</strong> the body to the exterior where it<br />

can be used for rapid response in the flight or fight response (as reviewed by Carroll and<br />

Forsberg, 2007). There is also a redistribution <strong>of</strong> leukocytes from various compartments resulting<br />

in increased numbers <strong>of</strong> white blood cells in peripheral circulation. This results in enhancement<br />

<strong>of</strong> the cell mediated immune response as a result <strong>of</strong> acute stress (Dhabhar and Mcewen, 1997). In<br />

the case <strong>of</strong> chronic stress, there is an increased duration at, which glucocorticoids act. This<br />

negates the positive aspects <strong>of</strong> acute stress on the immune system and results in suppression <strong>of</strong><br />

interleukin-6 and the cell mediated immune response. This is an adaptive response since repeated<br />

stimulation <strong>of</strong> the HPA axis and the immune system by glucocorticoids could lead to an<br />

increased risk <strong>of</strong> autoimmune disease (Dhabhar and Mcewen, 1997). Chronic stress can lead to<br />

decreased growth and performance in livestock (Barnett and Hemsworth, 1986; Hemsworth et<br />

al., 2000). Two <strong>of</strong> the main stressors for the milk-fed calf are nutritional stress and social<br />

isolation. Two <strong>of</strong> the main stressors for the post-weaned calf are social mixing and transport<br />

stress.<br />

9


1.4.2 Nutritional Stressors<br />

Dairy calves are primarily fed a milk based diet until weaning is initiated. Energy intake<br />

in the pre-weaning period is currently recommended to be 4.8-5.3 Mcal/day for a 50 kg calf<br />

under thermoneutral conditions in order to gain 1.0 kg/day (Drackley, 2008). The traditional<br />

feeding practice for North American calves is to feed 10% <strong>of</strong> their body weight; a 45 kg calf is<br />

fed 4.5 kg <strong>of</strong> milk. However, recent data from the National Research Council found that 45 kg<br />

calves need to consume approximately 2.5L <strong>of</strong> whole milk for maintenance and the remaining<br />

milk can be used for growth (National Research Council, 2001; Drackley, 2008). These<br />

calculations are based on a thermoneutral environment , which does not require animals to exert<br />

energy to maintain homeostasis; heat or cold can increase the maintenance demands <strong>of</strong> calves<br />

further reducing the energy available for growth on a restricted feeding regime (as reviewed by<br />

Drackley, 2008). For this reason, the recommended caloric intake <strong>of</strong> calves has begun to change,<br />

with the National Farm Animal Council’s Code <strong>of</strong> Practice for Care and Handling <strong>of</strong> Dairy<br />

Cattle recommending calves be fed 20% <strong>of</strong> their body weight (National Farm Animal Council,<br />

2009).<br />

To avoid a post-weaning slump in weight gain when calves transition from a primarily<br />

milk based diet to a grain based diet it is essential that their rumen is adequately developed to<br />

absorb nutrients from a grain based diet. At birth, a calf’s rumen is small and underdeveloped.<br />

Highly fermentable dry feed is necessary for it to develop in size, musculature and absorptive<br />

capacity (Davis and Drackley, 1998). It takes approximately three weeks <strong>of</strong> grain and water<br />

intake for the rumen to develop to a size and absorptive capacity that is necessary for<br />

maintenance <strong>of</strong> growth and four months for it to reach the size <strong>of</strong> a mature animal (Davis and<br />

10


Drackley, 1998). At the time <strong>of</strong> weaning, it is a general guideline that calves need to consume at<br />

least 0.7 to 0.9 kg <strong>of</strong> calf starter each day for three consecutive days in order to maintain a<br />

modest weight gain (Davis and Drackley, 1998). If calves are able to consume this volume their<br />

rumen should be adequately developed to sustain growth through the transition from a liquid<br />

milk based diet to a solid diet consisting <strong>of</strong> roughage and calf starter. Rumens can be adequately<br />

developed to digest grains and release volatile fatty acids at three weeks <strong>of</strong> age (Kehoe et al.,<br />

2007). Early weaning, defined as 5 weeks compared to 14 weeks <strong>of</strong> age, has been promoted for<br />

many years as a method to reduce the expenses associated with raising calves (Castle and<br />

Watson, 1959). This is primarily driven by the high cost <strong>of</strong> rearing calves prior to weaning<br />

(Gabler et al., 2000). However, this practice has been questioned on the grounds <strong>of</strong> animal<br />

welfare since early weaning and low milk volumes are not the natural feeding style <strong>of</strong> calves and<br />

hunger has been shown to occur in restricted fed calves (De Paula Veiera et al, 2008).<br />

Until recently, feeding high volumes <strong>of</strong> milk to replacement heifers was thought to be<br />

associated with decreased mammary development. While early published studies reported that<br />

increased growth is associated with decreased mammary development, this finding was restricted<br />

to heifers older than 3 months <strong>of</strong> age (as reviewed by Sejrsen, 1994). However, a lack <strong>of</strong><br />

association between a higher plane <strong>of</strong> nutrition and mammary development was first reported by<br />

Meyer and colleagues (2006) when age differences at euthanasia were controlled for. Recently, it<br />

has also been reported that increased pre-weaning growth is associated with improved mammary<br />

development and decreased age at first calving (Raeth-Knight et al., 2009). This contradicts<br />

previous research and suggests benefits for high volume milk feeding. An additional concern <strong>of</strong><br />

limit-feeding dairy calves is the higher energy demand required during a febrile response. In<br />

11


humans, the metabolic rate increases by 13% for every degree Celsius rise in temperature<br />

(Kluger, 1978; Hart, 1988). Calves that have already been through a period <strong>of</strong> poor nutrition due<br />

to restricted feed intake or poor rumen development prior to becoming ill are further<br />

disadvantaged since they have lower nutrient availability to fight infection (Carroll and Forsberg,<br />

2007).<br />

A challenge associated with increased nutrient intake pre-weaning, or ad libitum milk<br />

feeding programs, is the issue <strong>of</strong> transitioning calves onto solid feed. When calves are <strong>of</strong>fered ad<br />

libitum access to milk they will consume on average 10 L per day (Appleby et al., 2001). While<br />

feeding at this level improves calf welfare by reducing the hunger and frustration experienced<br />

during a limited feeding program (De Paula Vieira et al., 2008), it also reduces the motivation for<br />

calves to consume calf starter. Some <strong>of</strong> the methods to encourage starter intake and reduce<br />

weaning distress have included reducing the number <strong>of</strong> milk meals per day (Kehoe et al., 2007),<br />

dilution <strong>of</strong> milk, provision <strong>of</strong> warm water, extended access to the feeding apparatus (Budzynska<br />

and Weary, 2008), extended weaning periods (Sweeney et al., 2010), and rearing calves in pairs<br />

(De Paula Vieira et al., 2010). These methods have begun to address the issue <strong>of</strong> weaning calves<br />

from high milk volume feeding regimes, however, research in this area needs to be continued to<br />

counteract the negative effects <strong>of</strong> poor weaning on animal welfare and productivity. Given that<br />

there appear to be long-term consequences for the productivity <strong>of</strong> dairy calves based on rearing<br />

procedures, future economic analysis <strong>of</strong> early milk feeding programs should include these long-<br />

term effects and not just the cost <strong>of</strong> feeding.<br />

12


1.4.3 Social Stressors<br />

Another stressor occurring around weaning is social stress. In studies <strong>of</strong> mature cattle, it<br />

takes ten days for cattle to establish a social hierarchy (Tennessen et al., 1985). This is a stressful<br />

time for adult cattle and is accompanied by decreased milk yield on the day <strong>of</strong> mixing. It is<br />

therefore likely that the initial exposure and mixing <strong>of</strong> calves reared in isolation to new herd<br />

mates is a stressful experience.<br />

Social interactions in established groups are very important to cattle. Operant<br />

conditioning can be used to determine how much value an animal places on certain activities.<br />

Two-month-old calves are willing to work hard to access full social contact with known<br />

conspecifics (Boissy and Le Neindre, 1990; Holm et al., 2002). They still work for nose-to-nose<br />

contact, but place a higher value on full contact as shown by an increased willingness to work for<br />

the reward <strong>of</strong> social contact. Yearling heifers were trained to access a feed reward either in<br />

isolation, in the presence <strong>of</strong> conspecifics that they had a strong affinity to, or in the presence <strong>of</strong> a<br />

conspecific they had a weak affinity towards as measured by social contact prior to testing<br />

(Boissy and Le Neindre, 1990). In order to eliminate competition heifers only had visual contact<br />

with ‘spectator’ heifers during operant testing situations. Results indicate that heifers learn faster<br />

when in the presence <strong>of</strong> conspecifics, even if they are unable to physically interact with them.<br />

This result was most noticeable when heifers were in the presence <strong>of</strong> a spectator heifer they had<br />

strong affinity towards. However, heifers trained in the presence <strong>of</strong> an unfamiliar spectator<br />

animal still learned to access the food reward faster than heifers that were trained in isolation.<br />

These results indicate that social interactions are beneficial for cattle.<br />

13


Without human intervention, calves first bond with their dam. Le Neindre (1989) found<br />

that beef calves that are left with their dams until they are weaned had preferential attachment to<br />

their dams. If calves are allowed to remain with dams through the next calving, the bond between<br />

mother-dam is not altered (Veissier et al., 1990). However, due to the nature <strong>of</strong> the dairy<br />

industry, dairy calves do not have this initial bond formation with their dam and must form<br />

bonds with their conspecifics without this initial socialization.<br />

Positive Social Interactions<br />

Grooming, pair formation and play behaviour are three examples <strong>of</strong> positive social<br />

interactions that strengthen the bonds between animals within a group. Studies on herds <strong>of</strong> cattle<br />

indicate that grooming behavior is most likely to be directed at animals that are <strong>of</strong> similar social<br />

dominance rank. Frequency <strong>of</strong> grooming was positively associated with time spent in<br />

cohabitation (Sato, 1984; Sato et al., 1991). Environmental factors, such as delayed feeding time<br />

and dirty environments, also influence the time spent social licking. However, the preferential<br />

partnering and increased growth rates associated with animals that receive social grooming<br />

indicates that this likely functions as a bonding and relaxing activity for cattle, as well as for<br />

cleaning. The decreased heart rate <strong>of</strong> cattle that receive allogrooming supports the hypothesis<br />

that this is a rewarding behavior, which can improve cattle welfare (Laister et al., 2011). Since<br />

individually housed neonates are prevented from contacting conspecifics they are also prevented<br />

from performing and receiving this soothing activity, likely affecting their response to stress.<br />

Within groups <strong>of</strong> cattle preferential partnerships develop. In extensive and feral herds <strong>of</strong><br />

cattle it is initially the dam-young relationship that is given priority (Lazo, 1994). However, by<br />

two weeks <strong>of</strong> age calves spend the majority <strong>of</strong> their time with other calves, when nursing time is<br />

14


emoved from the time budget (Wood-Gush et al., 1984). This allows bonds to form between<br />

unrelated animals. When 9 male and 9 female prepubertal calves were studied within a large<br />

extensive herd where they had minimal human contact, four calf pairs were found. A pair <strong>of</strong><br />

calves was defined as two animals that mutually preferred to spend time with each other in each<br />

<strong>of</strong> the six months <strong>of</strong> this study. Interestingly, on average pair members differed in age by 5 days<br />

despite the fact that there was an age range <strong>of</strong> two months within the group (Reinhardt et al.,<br />

1978). This indicates that it is likely that age and developmental stage are very important in the<br />

development <strong>of</strong> partners within a herd; Reinhardt and colleagues (1978) concluded that social<br />

contact between calves is not random, but also highly unpredictable.<br />

Play behaviour is one aspect <strong>of</strong> social behavior. Calves move in and out <strong>of</strong> play behavior<br />

quite rapidly and under natural conditions perform this behavior many times throughout the day<br />

(Reinhardt et al, 1978). Once play is initiated it appears to elicit this behavior in conspecifics not<br />

involved in the original event (Vitale, 1986). Play behaviour occurs between both sexes with<br />

males being more active and more likely to be involved in social activity than the females in the<br />

group (Reinhardt et al., 1978; Vitale, 1986). There has been interest in using play behavior as an<br />

indicator <strong>of</strong> positive welfare. Play is predicted to be an indicator <strong>of</strong> positive welfare based on<br />

decreases in play behavior during aversive situations, such as hunger and pain. For example, play<br />

behavior is decreased in lambs following castration (Thornton and Waterman-Pearson, 2002) and<br />

in groups <strong>of</strong> dairy calves during weaning (Krachun et al., 2010).<br />

Negative Social Interactions<br />

In modern dairy systems, cattle are kept in larger groups than they would normally<br />

encounter in feral or range conditions. In the mountains <strong>of</strong> Japan, a feral herd <strong>of</strong> cattle had an<br />

15


average group size <strong>of</strong> three or four cattle and aggressive interactions were minimal except at<br />

water, the only limiting resource (Kimura and Ihobe, 1985). However, these animals had a very<br />

low-density range <strong>of</strong> approximately 10 individuals per km 2 . This is not a stocking density that is<br />

viable in the modern dairy industry and as such aggressive interactions are more common in the<br />

commercial dairy industry. Cattle have been shown to form a dominance hierarchy, which<br />

consists <strong>of</strong> a range <strong>of</strong> dominant and submissive animals in each group. This hierarchy is most<br />

commonly defined as a linear relationship for post-pubertal beef and dairy cattle (Schein and<br />

Fohrman, 1955; Stricklin et al., 1980; Stricklin, 1983).<br />

Benefits <strong>of</strong> being dominant include better access to resources such as feed, water and<br />

stalls. Researchers have found that dominant animals get more rest, can move more freely and<br />

eat first and for longer in comparison to submissive animals (Ol<strong>of</strong>sson, 1999). This benefits the<br />

dominant cattle by allowing them to better maintain body condition. The health status <strong>of</strong> cattle<br />

and dominance rank is closely linked, and likely feeds back on itself. A healthy cow is more<br />

likely to be dominant and subordinate cows are more likely to become ill. Galindo and Broom<br />

(2000) found that 28% <strong>of</strong> high ranking cows became lame during the study period versus 60% <strong>of</strong><br />

low ranking cows. In a free-stall management system, dominant cows are less likely to be lame,<br />

perhaps as a result <strong>of</strong> an ability to control resources such as stalls and feed bunks. This forces the<br />

subordinate animals to walk further and rest less, putting them at increased risk <strong>of</strong> ho<strong>of</strong><br />

problems. Agonistic encounters also cause chronic stress in subordinate animals, suppressing the<br />

immune system and resulting in increased susceptibility to other diseases (González et al., 2003).<br />

Older animals are more likely to be dominant over younger animals. This predictor is<br />

highly influenced by the experience and size <strong>of</strong> the older animal. Dickson et al (1970) found that<br />

16


age and body size are correlated with dominance. The ability to win a dominance challenge,<br />

however, appears to be highly heritable with genetics being a much larger factor in future<br />

dominance than rearing environment. Studies <strong>of</strong> mononzygotic twins show that when reared<br />

separately from the dam the dominance status <strong>of</strong> twins is highly correlated (r=0.93; Purcell and<br />

Arave, 1991). Agonistic behaviour develops later in life than affilitative behaviours. Agonistic<br />

interactions are rarely observed in pre-pubertal cattle kept in semi-wild conditions with a stable<br />

social structure (Reinhardt et al., 1978).<br />

It is not known if young calves are capable <strong>of</strong> establishing a dominance hierarchy through<br />

agonistic interactions. Studies conducted to investigate mixing <strong>of</strong> calves have produced very<br />

different results. When pairs <strong>of</strong> 11-month old Holstein heifers were mixed weekly over a period<br />

<strong>of</strong> 16 weeks it was found that aggressive interactions at the time <strong>of</strong> mixing remained at similar<br />

levels throughout the study period (Raussi et al., 2005). However, when pairs <strong>of</strong> 3-14 week old<br />

Montbeliard veal calves were repeatedly mixed, they did not display a significant stress<br />

response, and although they performed increased activity on the day <strong>of</strong> mixing, there was a low<br />

level <strong>of</strong> aggressive interactions (Veissier et al., 2001). This would suggest that age plays a role in<br />

the development <strong>of</strong> aggressive interactions. However, it has not been determined when exactly<br />

these interactions begin to occur and if they are initiated by physical, environmental or social<br />

changes in the calves. For example, milk fed calves will displace each other at the feeder, which<br />

indicates that competition for resources can occur in young animals (Jensen and Budde, 2006).<br />

This can be alleviated by altering either the ability <strong>of</strong> competitors to access the resource, such as<br />

the milk feeder (Jensen et al., 2008), or by altering the demand for the resource through<br />

increasing milk volume fed or the number <strong>of</strong> teats available (Jensen, 2004; von Keyserlingk et<br />

17


al., 2004). Group composition and group size also affect the level <strong>of</strong> aggression with a wider age<br />

variation associated with increased levels <strong>of</strong> aggression and younger animals most likely to be<br />

targeted (Færevik et al., 2010). A study comparing pair housed calves to calves housed in groups<br />

<strong>of</strong> 4 found that calves consumed milk faster in groups <strong>of</strong> four compared to groups <strong>of</strong> two,<br />

indicating increased competition in groups <strong>of</strong> four (Jensen and Budde, 2006). For groups <strong>of</strong> four,<br />

six and eight calves, feed bunk displacements after mixing were highest in groups <strong>of</strong> four, but<br />

there was no effect <strong>of</strong> group size on displacements from the lying area (Færevik et al., 2007).<br />

When examining established groups, group size was not found to be associated with increased<br />

aggression (Kondo et al., 1989). This may indicate that group size only affects agonistic<br />

interactions during mixing, when social hierarchy is being determined.<br />

In conclusion, social interactions can either increase or decrease the welfare <strong>of</strong> calves<br />

depending on their status within the group, group composition and group stability. In extensive<br />

groups, animals have the potential to join other herds, form different sub-groups and avoid<br />

dominant animals. In confined spaces these options are restricted or eliminated. For this reason,<br />

management <strong>of</strong> these social animals needs to focus on refining mixing practices and reducing<br />

competition in order to reduce the negative impacts <strong>of</strong> social life on low-ranking animals.<br />

1.4.4 Transport Stress<br />

It is common practice for calves to be moved to a novel environment post-weaning. This<br />

will require calves to either be restrained with a halter or loaded into a trailer in order to relocate<br />

them. This will result in acute or chronic stress depending on the duration <strong>of</strong> travel (Apple et al.,<br />

2005). In addition to the stress <strong>of</strong> travel, calves are taken from an isolated environment that limits<br />

disease exposure to being in contact with many heifers <strong>of</strong> varying age groups, and consequently<br />

18


exposed to many different disease causing organisms. When this stressor is combined with<br />

weaning, disease outbreaks are likely, as shown by the high risk <strong>of</strong> BRD post-weaning in both<br />

dairy calves (McGuirk, 2008) and beef feedlots (Smith, 1998). Efforts to minimize stress<br />

associated with transport to a novel environment are important for reducing weaning stress and<br />

minimizing the risk <strong>of</strong> disease.<br />

1.5 CONTROL OF DISEASE SPREAD<br />

At the herd level, management <strong>of</strong> BRD has included vaccination, removal <strong>of</strong> stressors,<br />

and administration <strong>of</strong> antibiotics in the water or feed, and prevention <strong>of</strong> other diseases.<br />

Vaccination is a challenge in the face <strong>of</strong> maternal antibodies received from colostrum; as such,<br />

vaccination <strong>of</strong> young dairy calves for respiratory disease is difficult (Ellis et al., 2001; Fulton et<br />

al., 2004). For this reason, disease prevention must continue to address management factors that<br />

affect the spread <strong>of</strong> disease, the stress levels <strong>of</strong> the calf and reduce the viral and bacterial load in<br />

the environment.<br />

Risk factors for BRD include concurrent nutritional, environmental and social changes,<br />

as well as increased pathogen exposure from mixing with older animals for the first time.<br />

Important control measures include vaccinating cows against the causal organisms, ensuring<br />

excellent colostrum management and minimizing acute and chronic stressors such as poor<br />

ventilation, mixing <strong>of</strong> groups, over-crowding and poor nutrition (Kahn, 2005). However, at<br />

weaning or during the change <strong>of</strong> housing after weaning some stressors cannot be avoided,<br />

resulting in increased incidence <strong>of</strong> this disease (McGuirk, 2008).<br />

One approach used by producers to reduce disease is the administration <strong>of</strong> antibiotics at<br />

the group level. There are currently two approaches for group administration <strong>of</strong> antibiotics. The<br />

19


first is the addition <strong>of</strong> antibiotics to the feed and the second is a single injection <strong>of</strong> antibiotic to<br />

all animals in a group at a specific time point. The most recent NAHMS survey reported that<br />

18% <strong>of</strong> dairy operations used an antibiotic in the feed <strong>of</strong> post-weaned heifers (USDA, 2009). The<br />

addition <strong>of</strong> antibiotics to the feed is a very convenient method to administer antibiotics to calves;<br />

it eliminates the cost <strong>of</strong> handling animals to administer an injection and when intake is monitored<br />

to ensure adequate intake, has been shown to successfully treat clinical disease (Fodor et al.,<br />

2000). The main drawback <strong>of</strong> this approach is the challenge <strong>of</strong> administering an appropriate dose<br />

to all animals, especially those at highest risk. The animals that are most likely to receive the<br />

highest dose are those with high feed consumption in the immediate post-weaning period. These<br />

animals are already less likely to become ill since high feed consumption reduces or eliminates<br />

the negative energy balance that can occur post-weaning. In contrast animals that are not<br />

consuming adequate feed, either due to poor adaptation to weaning or the initial stages <strong>of</strong><br />

disease, will get a lower dose <strong>of</strong> antibiotics despite a higher risk <strong>of</strong> disease.<br />

The main drawback <strong>of</strong> injecting antibiotics is the cost <strong>of</strong> administering the drugs and the<br />

shorter duration <strong>of</strong> activity. The labour costs to restrain and inject a large group <strong>of</strong> calves can be<br />

very high. Application <strong>of</strong> proper restraint is costly due to the time required and is likely stressful<br />

for the calves. The benefit <strong>of</strong> this approach is that calves will receive a correct dose based on<br />

their size with very little risk <strong>of</strong> under or over dosing. This approach was first used in the beef<br />

feedlot industry for the prevention <strong>of</strong> respiratory disease in animals expected to be at high risk <strong>of</strong><br />

morbidity and mortality. Animals considered to be at high risk are animals stressed due to the<br />

abrupt severing <strong>of</strong> the maternal bond, transportation, comingling, and marketing (Duff and<br />

Galyean, 2007). In 1999, 27% <strong>of</strong> small beef feedlots and 81% <strong>of</strong> large feedlots in the United<br />

20


States were administering an injectable antimicrobial metaphylactically to all cattle for the<br />

prevention <strong>of</strong> respiratory disease (USDA, 2000). A meta-analysis <strong>of</strong> field trials indicated that for<br />

feedlot calves, administration <strong>of</strong> oxytetracycline or tilmicosin on the day <strong>of</strong> arrival at the feedlot<br />

consistently reduced morbidity rates attributable to BRD (Van Donkersgoed et al., 1993). The<br />

movement <strong>of</strong> beef cattle into a feedlot is comparable to the movement <strong>of</strong> dairy calves from<br />

individual to group housing. In both cases, calves experience similar stressors, which include<br />

nutritional, environmental and social changes, transportation, and increased pathogen exposure.<br />

Regardless <strong>of</strong> the method used to administer antibiotics, this practice must be carefully<br />

evaluated by both researchers and producers to maximize efficacy and avoid unnecessary use<br />

given the concern over developing antimicrobial resistance. The ideal circumstances for adopting<br />

this prophylactic program are the following: a targeted disease that is highly contagious and is<br />

endemic to the region, successful clinical treatment is difficult, and the risk <strong>of</strong> pathogen exposure<br />

is constrained within a short time period. In conjunction with this method, efforts should be<br />

made to reduce or minimize manageable risk factors with the goal <strong>of</strong> reducing or eliminating the<br />

need for prophylactic use. For this program to be most effective good record keeping is essential<br />

in order to review historical disease trends to identify high risk time periods and environmental<br />

triggers. In addition, these records are needed for outcome assessment to measure the success <strong>of</strong><br />

this program and other management changes made to decrease the risk <strong>of</strong> disease for efficacy.<br />

1.6 LONG-TERM IMPACTS OF BOVINE RESPIRATORY DISEASE COMPLEX<br />

BRD is a complex <strong>of</strong> disease processes most commonly resulting from infection with<br />

various microorganisms, including Pasteurella multocida, Mannheimia haemolytica and<br />

Mycoplasma bovis, Histophilus somni in conjunction with bovine herpesvirus-1 (BHV-1) bovine<br />

21


viral diarrhea virus (BVDV) bovine respiratory syncytial virus (BRSV) and parainfluenza 3 virus<br />

(PI3V) (Kahn, 2005). The development <strong>of</strong> BRD is typically initiated by an environmental<br />

stressor and viral infection, which weakens the resistance mechanisms <strong>of</strong> the lungs and allows<br />

bacterial colonization (Yates, 1984). There are many factors that have been associated with<br />

increased risk <strong>of</strong> BRD, including mixing <strong>of</strong> animals, transportation, light weight or younger<br />

animals within a group and persistent infection with BVDV (as reviewed by Taylor et al., 2010).<br />

The National Animal Health Monitoring Survey (NAHMS) dairy heifer study found that 6% <strong>of</strong><br />

weaned dairy heifers were treated in 2007 for respiratory disease (USDA, 2009). However, the<br />

NAHMS study defined post-weaned heifers as including all animals from weaning to first<br />

calving. The wide age variation encompassed by this definition would include many older<br />

animals that were at low risk <strong>of</strong> BRD. Other epidemiologic studies in North America have<br />

reported BRD rates in milk fed dairy calves ranging from 0.1 cases/100 days in Minnesota<br />

(Sivula et al., 1996), 25% <strong>of</strong> calves in New York State (Virtala et al., 1996) and 39% <strong>of</strong> calves in<br />

Saskatchewan (Van Donkersgoed et al., 1993). This substantial variation in the reported rates <strong>of</strong><br />

disease probably results from a combination <strong>of</strong> the differences in the diagnostic criteria used,<br />

disease definition, age <strong>of</strong> the calves, the composition <strong>of</strong> the viral and bacterial agents, and<br />

management conditions.<br />

Costs <strong>of</strong> BRD were estimated at $14.71(range: 0 - $119) per pre-weaned calf per year and<br />

$1.95 (range: 0-$9.25) per weaned dairy calf stock per year by Michigan producers in 1990<br />

(Kaneene and Hurd, 1990). These costs included labor for detection and treatment <strong>of</strong> disease,<br />

drugs, veterinary fees and cost to replace animals that are died as a result <strong>of</strong> clinical disease. The<br />

higher cost observed in young calves is based on the timing <strong>of</strong> preventive practices, which<br />

22


primarily are administered to calves prior to disease outbreaks, and the long follow-up time for<br />

calf stock, which included an extended period <strong>of</strong> low disease risk. However, these estimates<br />

likely underestimate the cost <strong>of</strong> BRD due to the difficulty in associating a disease event with<br />

outcomes many months later. The European estimate <strong>of</strong> 18-57 euros ($ 25-81 U.S.) per case is<br />

likely more accurate since the effect <strong>of</strong> BRD on growth, long-term survival and productivity<br />

were considered (van der Fels-Klerx et al., 2001). However, the effects <strong>of</strong> BRD on growth,<br />

survival and production were based on a small number <strong>of</strong> studies examining these factors that<br />

were conducted 10-20 years ago. Results from these three studies found that BRD increased the<br />

risk <strong>of</strong> mortality prior to calving (Waltner-Toews et al., 1986), decreased growth (Virtala et al.,<br />

1996), increased age at calving (Waltner-Toews et al., 1986; Correa et al., 1988) and increased<br />

the risk <strong>of</strong> dystocia at first calving(Warnick et al., 1994). Warnick et al, (1995; 1997) found no<br />

association <strong>of</strong> calfhood BRD with first lactation milk production or survival after first calving.<br />

The welfare costs <strong>of</strong> BRD can be substantial based on both the clinical state and the long-<br />

term implications <strong>of</strong> this disease. The clinical signs can include inappetance, depression,<br />

coughing and shortness <strong>of</strong> breath (dyspnea), which compromise animal welfare (Mellor and<br />

Stafford, 2004; Kahn, 2005). The long-term effects <strong>of</strong> BRD on body weight gain and survival are<br />

<strong>of</strong> concern since these persistent consequences have the potential to decrease calf welfare.<br />

Specifically, decreased growth may be due to a lower dominance status leading to decreased<br />

ability to access resources or an ongoing disease challenge, which would be associated with<br />

ongoing clinical signs. In both <strong>of</strong> these scenarios animal welfare would be affected and new<br />

management techniques and treatments for affected animals are necessary to improve<br />

convalescence in these animals and to reduce the long-term impacts <strong>of</strong> this disease.<br />

23


1.7 CONCLUSION AND THESIS OBJECTIVES<br />

In conclusion, the short and long term impacts <strong>of</strong> dairy calf management on their welfare<br />

needs to be evaluated, within the confines <strong>of</strong> a viable economic model. Currently, there are many<br />

gaps in our knowledge and there may be ways to improve management systems, including<br />

identifying behaviors that can be used for early identification <strong>of</strong> disease, and understanding some<br />

<strong>of</strong> the long-term impacts <strong>of</strong> disease. The goals <strong>of</strong> this thesis were to investigate the use <strong>of</strong><br />

behavior and activity monitoring for the identification <strong>of</strong> animals at risk <strong>of</strong> disease, the use <strong>of</strong><br />

group level management practices to reduce the risk <strong>of</strong> neonatal calf diseases, and the<br />

identification <strong>of</strong> long-term impacts <strong>of</strong> respiratory disease.<br />

Specifically the research objectives were;<br />

1) To identify behaviors associated with poor weight gain in newly weaned dairy calves.<br />

2) To evaluate the impact <strong>of</strong> a novel management practice to reduce stress associated<br />

with transitioning weaned calves to a novel environment.<br />

3) To identify behaviors associated with disease and poor growth in milk-fed dairy calves<br />

4) To evaluate the impact <strong>of</strong> a targeted antibiotic approach to reduce disease in milk-fed<br />

dairy calves<br />

5 To evaluate the efficacy <strong>of</strong> tulathromycin on the incidence <strong>of</strong> BRD and growth in<br />

weaned calves following movement to group housing in a novel environment.<br />

6) To evaluate the long-term impacts <strong>of</strong> BRD on the welfare and productivity <strong>of</strong> dairy<br />

cattle.<br />

24


1.8 REFERENCES<br />

Al-Chaer, E. D. and R. J. Traub. 2002. Biological basis <strong>of</strong> visceral pain: Recent developments.<br />

Pain. 96:221-225.<br />

Apple, J. K., E. B. Kegley, D. L. Galloway, T. J. Wistuba and L. K. Rakes. 2005. Duration <strong>of</strong><br />

restraint and isolation stress as a model to study the dark-cutting condition in cattle. J. Anim. Sci.<br />

83:1202-1214.<br />

Appleby, M. C., D. M. Weary and B. Chua. 2001. Performance and feeding behaviour <strong>of</strong> calves<br />

on ad libitum milk from artificial teats. Appl. Anim. Behav. Sci. 74:191-201.<br />

Aubert, A. 1999. Sickness and behaviour in animals: A motivational perspective. Neuroscience<br />

& Biobehavioral Reviews. 23:1029-1036.<br />

Avitsur, R. and R. Yirmiya. 1999. The immunobiology <strong>of</strong> sexual behavior: Gender differences in<br />

the suppression <strong>of</strong> sexual activity during illness. Pharmacology Biochemistry and Behavior.<br />

64:787-796.<br />

Barnett, J. and P. Hemsworth. 1986. The impact <strong>of</strong> handling and environmental factors on the<br />

stress response and its consequences in swine. Lab. Anim. Sci. 36:366-369.<br />

Bluthé, R., M. Pawlowski, S. Suarez, P. Parnet, Q. Pittman, K. Kelley and R. Dantzer. 1994.<br />

Synergy between tumor necrosis factor alpha and interleukin-1 in the induction <strong>of</strong> sickness<br />

behavior in mice. Psychoneuroendocrinology. 19:197-207.<br />

Bluthe´, R., R. Dantzer and K. W. Kelley. 1992. Effects <strong>of</strong> interleukin-1 receptor antagonist on<br />

the behavioral effects <strong>of</strong> lipopolysaccharide in rat. Brain Research. 573:318-320.<br />

Boissy, A. and P. Le Neindre. 1990. Social influences on the reactivity <strong>of</strong> heifers: Implications<br />

for learning abilities in operant conditioning. Appl. Anim. Behav. Sci. 25:149-165.<br />

Borderas, T. F., A. M. B. de Passille and J. Rushen. 2008. Behavior <strong>of</strong> dairy calves following a<br />

low dose <strong>of</strong> bacterial endotoxin. J. Anim. Sci. 86: 2920-2927.<br />

Borderas, T. F., J. Rushen, M. A. G. von Keyserlingk and A. M. B. de Passille. 2009. Automated<br />

measurement <strong>of</strong> changes in feeding behavior <strong>of</strong> milk-fed calves associated with illness. J. Dairy<br />

Sci. 92:4549-4554.<br />

Budzynska, M. and D. M. Weary. 2008. Weaning distress in dairy calves: Effects <strong>of</strong> alternative<br />

weaning procedures. Applied Animal Behaviour Science. 112:33-39.<br />

25


Buhman, M. J., L. J. Perino, M. L. Galyean, T. E. Wittum, T. H. Montgomery and R. S. Swingle.<br />

2000. Association between changes in eating and drinking behaviors and respiratory tract disease<br />

in newly arrived calves at a feedlot. Am. J. Vet. Res. 61:1163-1168.<br />

Carroll, J. and N. E. Forsberg. 2007. Influence <strong>of</strong> stress and nutrition on cattle immunity. Vet<br />

Clin Food Anim. 23:105-149.<br />

Castle, M. E. and J. N. Watson. 1959. A comparison between an early weaning and a more<br />

conventional system <strong>of</strong> rearing dairy calves. Animal Production. 1:31-36.<br />

Cohn, D. W. H. and L. C. de Sá-Rocha. 2006. Differential effects <strong>of</strong> lipopolysaccharide in the<br />

social behavior <strong>of</strong> dominant and submissive mice. Physiol. Behav. 87:932-937.<br />

Correa, M. T., C. R. Curtis, H. N. Erb and M. E. White. 1988. Effect <strong>of</strong> calfhood morbidity on<br />

age at first calving in New York herds. Prev. Vet. Med. 6:253-262.<br />

Davis, C. L. and J. Drackley. 1998. The Development, Nutrition, and Management <strong>of</strong> the Young<br />

Calf. 1st ed. Iowa State <strong>University</strong> Press, Iowa.<br />

de la Fuente, R., M. Luzón, J. A. Ruiz-Santa-Quiteria, A. García, D. Cid, J. A. Orden, S. García,<br />

R. Sanz and M. Gómez-Bautista. 1999. Cryptosporidium and concurrent infections with other<br />

major enterophatogens in 1 to 30-day-old diarrheic dairy calves in central Spain. Vet. Parasitol.<br />

80:179-185.<br />

De Paula Vieira, A., M. A. G. von Keyserlingk and D. M. Weary. 2010. Effects <strong>of</strong> pair versus<br />

single housing on performance and behavior <strong>of</strong> dairy calves before and after weaning from milk.<br />

J. Dairy Sci. 93:3079-3085.<br />

De Paula Vieira, A., V. Guesdon, A. M. de Passillé, M. A. G. von Keyserlingk and D. M. Weary.<br />

2008. Behavioural indicators <strong>of</strong> hunger in dairy calves. Applied Animal Behaviour Science.<br />

109:180-189.<br />

Dhabhar, F. S. and B. S. Mcewen. 1997. Acute stress enhances while chronic stress suppresses<br />

immune function in vivo: A potential role for leukocyte trafficking. Brain Behav. Immun.<br />

11:286-306.<br />

Dhabhar, F. S. 2002. Stress-induced augmentation <strong>of</strong> immune function—The role <strong>of</strong> stress<br />

hormones, leukocyte trafficking, and cytokines. Brain Behav. Immun. 16:785-798.<br />

Dhabhar, F. S. and B. S. Mcewen. 1997. Acute stress enhances while chronic stress suppresses<br />

cell-mediated immunityin vivo:A potential role for leukocyte trafficking. Brain Behav. Immun.<br />

11:286-306.<br />

Drackley, J. K. 2008. Calf nutrition from birth to breeding. Vet. Clin. North Am. Food Anim.<br />

Pract. 24(1): 55-86.<br />

26


Duff, G. C. and M. L. Galyean. 2007. Board-invited review: Recent advances in management <strong>of</strong><br />

highly stressed, newly received feedlot cattle. J. Anim. Sci. 85:823-840.<br />

Ellis, J., K. West, V. Cortese, C. Konoby and D. Weigel. 2001. Effect <strong>of</strong> maternal antibodies on<br />

induction and persistence <strong>of</strong> vaccine-induced immune responses against bovine viral diarrhea<br />

virus type II in young calves. J. Am. Vet. Med. Assoc. 219:351-356.<br />

Færevik, G., M. B. Jensen and K. E. Bøe. 2010. The effect <strong>of</strong> group composition and age on<br />

social behavior and competition in groups <strong>of</strong> weaned dairy calves. J. Dairy Sci. 93:4274-4279.<br />

Færevik, G., I. L. Andersen, M. B. Jensen and K. E. Bøe. 2007. Increased group size reduces<br />

conflicts and strengthens the preference for familiar group mates after regrouping <strong>of</strong> weaned<br />

dairy calves (bos taurus). Appl. Anim Behav. Sci. 108:215-228.<br />

Fodor, L., L. Reeve-Johnson, A. Hodge and J. Varga. 2000. Efficacy evaluations <strong>of</strong> the use <strong>of</strong><br />

oral tilmicosin in pneumonic calves. The Veterinary Journal. 159:194-200.<br />

Fraser, D., D. M. Weary, E. A. Pajor and B. N. Milligan. 1997. A scientific conception <strong>of</strong> animal<br />

welfare that reflects ethical concerns. Animal Welfare. 6:187-205.<br />

Fulton, R. W., R. E. Briggs, M. E. Payton, A. W. Confer, J. T. Saliki, J. F. Ridpath, L. J. Burge<br />

and G. C. Duff. 2004. Maternally derived humoral immunity to bovine viral diarrhea virus<br />

(BVDV) 1a, BVDV1b, BVDV2, bovine herpesvirus-1, parainfluenza-3 virus bovine respiratory<br />

syncytial virus, mannheimia haemolytica and pasteurella multocida in beef calves, antibody<br />

decline by half-life studies and effect on response to vaccination. Vaccine. 22:644-650.<br />

Gabler, M. T., P. R. Tozer and A. J. Heinrichs. 2000. Development <strong>of</strong> a cost analysis spreadsheet<br />

for calculating the costs to raise a replacement dairy heifer. Journal <strong>of</strong> Dairy Science. 83:1104.<br />

Galindo, F. and D. M. Broom. 2000. The relationships between social behaviour <strong>of</strong> dairy cows<br />

and the occurrence <strong>of</strong> lameness in three herds. Research in Veterinary Science. 69:75-79.<br />

Goldhawk, C., N. Chapinal, D. M. Veira, D. M. Weary and M. A. G. von Keyserlingk. 2009.<br />

Prepartum feeding behavior is an early indicator <strong>of</strong> subclinical ketosis. J. Dairy Sci. 92:4971-<br />

4977.<br />

González, M., A. K. Yabuta and F. Galindo. 2003. Behaviour and adrenal activity <strong>of</strong> first<br />

parturition and multiparous cows under a competitive situation. Appl. Anim. Behav. Sci. 83:259-<br />

266.<br />

Hart, B. L. 1988. Biological basis <strong>of</strong> the behavior <strong>of</strong> sick animals. Neuroscience and<br />

Biobehavioral Reviews. 12:123-137.<br />

27


Hemsworth, P. H., G. J. Coleman, J. L. Barnett and S. Borg. 2000. Relationships between<br />

human-animal interactions and productivity <strong>of</strong> commercial dairy cows. J. Anim Sci. 78:2821-<br />

2831.<br />

Holm, L., M. B. Jensen and L. L. Jeppesen. 2002. Calves’ motivation for access to two different<br />

types <strong>of</strong> social contact measured by operant conditioning. Appl. Anim Behav. Sci. 79:175-194.<br />

Huzzey, J. M., D. M. Veira, D. M. Weary and M. A. G. von Keyserlingk. 2007. Prepartum<br />

behavior and dry matter intake identify dairy cows at risk for metritis. J. Dairy Sci. 90:3220-<br />

3233.<br />

Jensen, M. B. 2004. Computer-controlled milk feeding <strong>of</strong> dairy calves: The effects <strong>of</strong> number <strong>of</strong><br />

calves per feeder and number <strong>of</strong> milk portions on use <strong>of</strong> feeder and social behavior. J. Dairy Sci.<br />

87:3428-3438.<br />

Jensen, M. B. and M. Budde. 2006. The effects <strong>of</strong> milk feeding method and group size on<br />

feeding behavior and cross-sucking in group-housed dairy calves. J. Dairy Sci. 89:4778-4783.<br />

Jensen, M. B., A. M. de Passillé, M. A. G. von Keyserlingk and J. Rushen. 2008. A barrier can<br />

reduce competition over teats in pair-housed milk-fed calves. J. Dairy Sci. 91:1607-1613.<br />

Kahn, C. M. 2005. Merck Veterinary Manual. 9th ed. Merck, Whitehouse Station, N.J.<br />

Kaneene, J. B. and H. S. Hurd. 1990. The national animal health monitoring system in Michigan.<br />

III. cost estimates <strong>of</strong> selected dairy cattle diseases. Preventive Veterinary Medicine. 8:127-140.<br />

Kehoe, S. I., C. D. Dechow and A. J. Heinrichs. 2007. Effects <strong>of</strong> weaning age and milk feeding<br />

frequency on dairy calf growth, health and rumen parameters. Livestock Science. 110:267-272.<br />

Kimura, D. and H. Ihobe. 1985. Feral cattle (bos taurus) on Kuchinoshima island, southwestern<br />

Japan: Their stable ranging and unstable grouping. J. Ethology. 3:39-47.<br />

Kluger, M. 1978. The evolution and adaptive value <strong>of</strong> fever. Am. Sci. 66:38-43.<br />

Kondo, S., J. Sekine, M. Okubo and Y. Asahida. 1989. The effect <strong>of</strong> group size and space<br />

allowance on the agonistic and spacing behavior <strong>of</strong> cattle. Appl. Anim. Behav. Sci. 24:127-135.<br />

Krachun, C., J. Rushen and A. M. de Passillé. 2010. Play behaviour in dairy calves is reduced by<br />

weaning and by a low energy intake. Appl. Anim. Behav. Sci. 122:71-76.<br />

Laister, S., B. Stockinger, A. Regner, K. Zenger, U. Knierim and C. Winckler. 2011. Social<br />

licking in dairy cattle—Effects on heart rate in performers and receivers. Appl. Anim. Behav.<br />

Sci. 130:81-90.<br />

28


Lazo, A. 1994. Social segregation and the maintenance <strong>of</strong> social stability in a feral cattle<br />

population. Animal Behaviour. 48:1133-1141.<br />

McGuirk, S. M. 2008. Disease management <strong>of</strong> dairy calves and heifers. Vet. Clin. Food. Anim.<br />

24:139-153.<br />

Mellor, D. J. and K. J. Stafford. 2004. Animal welfare implications <strong>of</strong> neonatal mortality and<br />

morbidity in farm animals. Veterinary Journal. 168:118-133.<br />

National Research Council. 2001. Nutrient Requirements <strong>of</strong> Dairy Cattle; 7th ed. National<br />

Academy Press, Washington, DC.<br />

Ol<strong>of</strong>sson, J. 1999. Competition for total mixed diest fed for ad libitum intake using one or four<br />

cows per feeding station. J. Dairy Sci. 82:69-79.<br />

Owen-Ashley, N. T. and J. C. Wingfield. 2006. Seasonal modulation <strong>of</strong> sickness behavior in<br />

free-living northwestern song sparrows (melospiza melodia morphna). Journal <strong>of</strong> Experimental<br />

Biology. 209:3062-3070.<br />

Proudfoot, K. L., J. M. Huzzey and M. A. G. von Keyserlingk. 2009. The effect <strong>of</strong> dystocia on<br />

the dry matter intake and behavior <strong>of</strong> Holstein cows. J. Dairy Sci. 92:4937-4944.<br />

Purcell, D. and C. W. Arave. 1991. Isolation vs. group rearing in monozygous twin heifer calves.<br />

Appl. Anim. Behav. Sci. 31:147-156.<br />

Raeth-Knight, M., H. Chester-Jones, S. Hayes, J. Linn, R. Larson, D. Ziegler, B. Ziegler and N.<br />

Broadwater. 2009. Impact <strong>of</strong> conventional or intensive milk replacer programs on Holstein heifer<br />

performance through six months <strong>of</strong> age and during first lactation. J. Dairy Sci. 92:799-809.<br />

Raussi, S., A. Boissy, E. Delval, P. Pradel, J. Kaihilahti and I. Veissier. 2005. Does repeated<br />

regrouping alter the social behaviour <strong>of</strong> heifers? Appl. Anim Behav. Sci. 93:1-12.<br />

Reinhardt, V., F. M. Mutiso and A. Reinhardt. 1978. Social behaviour and social relationships<br />

between female and male prepubertal bovine calves (bos indicus). Appl. Anim. Ethol. 4:43-54.<br />

Sato, S. 1984. Social licking pattern and its relationship to social dominance and live weight gain<br />

in weaned calves. Appl. Anim. Behav. Sci. 12:25-32.<br />

Sato, S., S. Sako and A. Maeda. 1991. Social licking patterns in cattle (bos taurus): Influence <strong>of</strong><br />

environmental and social factors. Appl. Anim Behav. Sci. 32:3-12.<br />

Schaefer, A. L., N. J. Cook, J. S. Church, J. Basarab, B. Perry, C. Miller and A. K. W. Tong.<br />

2007. The use <strong>of</strong> infrared thermography as an early indicator <strong>of</strong> bovine respiratory disease<br />

complex in calves. Res. Vet. Sci. 83:376-384.<br />

29


Schein, M. W. and M. H. Fohrman. 1955. Social dominance relationships in a herd <strong>of</strong> dairy<br />

cattle. The British Journal <strong>of</strong> Animal Behaviour. 3:45-55.<br />

Sejrsen, K. 1994. Relationships between nutrition, puberty and mammary development in cattle.<br />

Proc. Nutr. Soc. 53:103.<br />

Sivula, N. J., T. R. Ames, W. E. Marsh and R. E. Werdin. 1996. Descriptive epidemiology <strong>of</strong><br />

morbidity and mortality in Minnesota dairy heifer calves. Prev. Vet. Med. 27:155-171.<br />

Smith, R. A. 1998. Impact <strong>of</strong> disease on feedlot performance: A review. J. Anim. Sci. 76:272-<br />

274.<br />

Snodgrass, D., H. Terzolo, D. Sherwood, I. Campbell, J. Menzies and B. Synge. 1986. Aetiology<br />

<strong>of</strong> diarrhoea in young calves. Veterinary Record. 119:31-34.<br />

Stricklin, . 1983. Matrilinear social dominance and spatial relationships among Angus and<br />

Hereford cows. Journal <strong>of</strong> Animal Science. 57:1397-1405.<br />

Stricklin, W. R., H. B. Graves, L. L. Wilson and R. K. Singh. 1980. Social organization among<br />

young beef cattle in confinement. Applied Animal Ethology. 6:211-219.<br />

Svensson, C. and J. Hultgren. 2008. Associations between housing, management, and morbidity<br />

during rearing and subsequent first-lactation milk production <strong>of</strong> dairy cows in southwest<br />

Sweden. J. Dairy Sci. 91:1510-1518.<br />

Svensson, C. and M. B. Jensen. 2007. Short communication: Identification <strong>of</strong> diseased calves by<br />

use <strong>of</strong> data from automatic milk feeders. J. Dairy Sci. 90:994-997.<br />

Sweeney, B. C., J. Rushen, D. M. Weary and A. M. de Passillé. 2010. Duration <strong>of</strong> weaning,<br />

starter intake, and weight gain <strong>of</strong> dairy calves fed large amounts <strong>of</strong> milk. J. Dairy Sci. 93:148-<br />

152.<br />

Taylor, J. D., R. W. Fulton, T. W. Lehenbauer, D. L. Step and A. W. Confer. 2010. The<br />

epidemiology <strong>of</strong> bovine respiratory disease: What is the evidence for predisposing factors? Can<br />

Vet J. 51:1095-1100.<br />

Tennessen, T., M. A. Price and R. Berg. 1985. The social interactions <strong>of</strong> young bulls and steers<br />

after regrouping. Appl. Anim. Behav. Sci. 14:37-47.<br />

Thornton, P. D. and A. E. Waterman-Pearson. 2002. Behavioural responses to castration in<br />

lambs. Anim. Welf. 11:203-212.<br />

Todd, C. G. 2007. An Evaluation <strong>of</strong> Meloxicam as a Supportive Therapy for Neonatal Calf<br />

Diarrhea Complex. MSc, <strong>University</strong> <strong>of</strong> <strong>Guelph</strong>, <strong>Guelph</strong>, Ontario.<br />

30


Tsigos, C. and G. Chrousos. 2002. Hypothalamic-pituitary-adrenal axis, neuroendocrine factors<br />

and stress. J. Psychosom. 53:865-871.<br />

USDA. 2010a. Beef 2007–08, Part V: Reference <strong>of</strong> Beef Cow-Calf Management Practices in the<br />

United States. USDA–APHIS–VS,CEAH., Fort Collins, CO. #532.0410.<br />

USDA. 2010b. Dairy 2007, Heifer Calf Health and Management Practices on U.S. Dairy<br />

Operations, 2007. USDA:APHIS:VS, CEAH., Fort Collins, CO. #550.0110.<br />

USDA. 2009. Dairy 2007, Part IV: Reference <strong>of</strong> Dairy Cattle Health and Management Practices<br />

in the United States, 2007. 2007th ed. USDA:APHIS:VS, CEAH., Fort Collins, CO.<br />

#N494.0209.<br />

USDA. 2007. Dairy 2007, Part I: Reference <strong>of</strong> Dairy Cattle Health and Management Practices in<br />

the United States. 2007 USDA-APHIS-VS, CEAH., Fort Collins, CO. #N480.1007.<br />

USDA. 2000. Part III: Health management and biosecurity in U.S. feedlots, 1999.<br />

USDA:APHIS:VS, CEAH, National Animal Health Monitoring System. Fort Collins, CO.<br />

#N336.1200.<br />

van der Fels-Klerx, H. J., J. T. Sorensen, A. W. Jalvingh and R. B. Huirne. 2001. An economic<br />

model to calculate farm-specific losses due to bovine respiratory disease in dairy heifers. Prev.<br />

Vet. Med. 51:75-94.<br />

Van Donkersgoed, J., C. Ribble, L. G. Boyer and H. Townsend. 1993. Epidemiological study <strong>of</strong><br />

enzootic pneumonia in dairy calves in Saskatchewan. Can J Vet Res. 57:247-254.<br />

Veissier, I., A. Boissy, A. M. dePassille, J. Rushen, C. G. van Reenen, S. Roussel, S. Andanson<br />

and P. Pradel. 2001. Calves' responses to repeated social regrouping and relocation. J. Anim. Sci.<br />

79:2580-2593.<br />

Veissier, I., D. Lamy and P. Le Neindre. 1990. Social behaviour in domestic beef cattle when<br />

yearling calves are left with the cows for the next calving. Appl. Anim. Behav. Sci. 27:193-200.<br />

Virtala, A. M. K., G. D. Mechor and Y. T. Grohn. 1996. The effect <strong>of</strong> calfhood diseases on<br />

growth <strong>of</strong> female dairy calves during the first 3 months <strong>of</strong> life in New York State. Journal <strong>of</strong><br />

Dairy Science. 79:1040-1049.<br />

Vitale, . 1986. Social behaviour <strong>of</strong> the calves <strong>of</strong> semi-wild maremma cattle, bos primigenius<br />

taurus. Applied Animal Behaviour Science. 16:217.<br />

von Keyserlingk, M. A. G., L. Brusius and D. M. Weary. 2004. Competition for teats and<br />

feeding behavior by group-housed dairy calves. J. Dairy Sci. 87:4190-4194.<br />

31


Waltner-Toews, D., S. W. Martin and A. H. Meek. 1986. The effect <strong>of</strong> early calfhood health<br />

status on survivorship and age at first calving. Can. J. Vet. Res. 50:314-317.<br />

Warnick, L. D., H. N. Erb and M. E. White. 1994. The association <strong>of</strong> calfhood morbidity with<br />

first-lactation calving age and dystocia in new york holstein herds. Kenya Vet. 18:177-179.<br />

Weary, D. M., J. M. Huzzey and M. A. G. von Keyserlingk. 2009. BOARD-INVITED<br />

REVIEW: Using behavior to predict and identify ill health in animals. J. Anim. Sci. 87:770-777.<br />

Wood-Gush, D. G. M., K. Hunt, K. Carson and S. G. C. Dennison. 1984. The early behaviour <strong>of</strong><br />

suckler calves in the field. Biol. Behav. 9:295-306.<br />

Yates, W. D. 1984. A review <strong>of</strong> infectious bovine rhinotracheitis, shipping fever pneumonia and<br />

viral-bacterial synergism in respiratory disease <strong>of</strong> cattle. Can J Comp Med. 46:225-263.<br />

32


CHAPTER 2: PREVALENCE OF SICKNESS BEHAVIOR IN NEONATAL DAIRY<br />

CALVES AT A COMMERCIAL HEIFER FACILITY AND ASSOCIATIONS WITH<br />

AVERAGE DAILY GAIN AND ILLNESS<br />

2.1 INTRODUCTION<br />

Across species there are sets <strong>of</strong> predictable physiologic and behavioral responses<br />

associated with pathogens, and parasitic infection. These responses include pyrexia, anorexia,<br />

depression, lethargy, and decreased grooming and are considered to have evolved because <strong>of</strong> the<br />

increased probability for survival <strong>of</strong> the host (Hart 1988). In cattle, these changes have been used<br />

to identify sick animals. There is interest in using these behaviors to identify and treat animals<br />

early in order to avoid the negative effects <strong>of</strong> disease on calves’ performance and welfare<br />

(Millman, 2007; Weary et al. 2009).<br />

Researchers have begun to evaluate technology capable <strong>of</strong> automatically monitoring and<br />

reporting changes in body temperature and behavior that will allow early identification <strong>of</strong> disease<br />

in animals. Some examples include the use <strong>of</strong> infrared thermography to detect increased body<br />

temperature (Schaefer et al., 2007) and monitoring <strong>of</strong> eating and drinking behavior (Buhman et<br />

al. 2000; Schaefer et al., 2007) in beef calves for the early detection <strong>of</strong> bovine respiratory<br />

disease. In dairy cattle changes in feeding, drinking, lying and social behavior have been used as<br />

early behavioral indicators in cows with a variety <strong>of</strong> transition diseases including metritis<br />

(Huzzey et al., 2007), dystocia (Proudfoot et al., 2009), and subclinical ketosis (Huzzey et al.<br />

2007; Goldhawk et al. 2009; Proudfoot et al., 2009).<br />

Changes in the feeding behavior <strong>of</strong> dairy calves have been used to identify sick calves.<br />

However, these studies have utilized automated feeders (Svensson & Jensen. 2007) or extended<br />

33


ehavioural observations ( Borderas, et al. 2008; Borderas et al. 2009), which are not feasible for<br />

many producers. Currently, the primary housing system for milk-fed calves in the United States<br />

is individual housing, with 75% <strong>of</strong> pre-weaned calves housed individually (USDA 2007).<br />

Milk-fed dairy calves, regardless <strong>of</strong> housing system, are highly susceptible to disease and<br />

it is important to have multiple, reliable indicators for disease so early diagnosis can occur. The<br />

two most important diseases in pre-weaned dairy calves in the United States are diarrhea and<br />

respiratory disease (USDA 2009). Neonatal calf diarrhea complex is a broad category <strong>of</strong> diseases<br />

which occurs in approximately 24% <strong>of</strong> dairy calves in the United States (USDA 2010). The main<br />

diarrheal agents in dairy calves are enterotoxigenic Escherichia coli K99, Salmonella spp.,<br />

bovine rotavirus, bovine coronavirus, and Cryptosporidum parvum, which may act<br />

independently or concurrently (Snodgrass et al. 1986; de la Fuente et al. 1999). The primary sign<br />

<strong>of</strong> this disease complex is diarrhea and the primary treatment for mild cases is supportive fluid<br />

replacement therapy. These pathogens can cause temporary changes and damage to the intestinal<br />

tract through destruction and sloughing <strong>of</strong> enterocytes, villous atrophy, and inflammation <strong>of</strong> the<br />

submucosa, causing inflammation and sensitization <strong>of</strong> the visceral afferent nerves. In people,<br />

these changes to the intestinal tract are associated with abdominal pain and discomfort (Al-Chaer<br />

& Traub, 2002).<br />

Recognition <strong>of</strong> key behaviors associated with common diseases and poor growth can<br />

allow new and experienced animal caretakers to improve identification <strong>of</strong> potentially sick<br />

animals. The objectives <strong>of</strong> this research project were to 1) determine the proportion <strong>of</strong> calves<br />

performing five behavioral measures related to sickness behavior at different ages, 2) determine<br />

which behavioral measures were associated with reduced growth and 3) determine if these<br />

34


ehaviors can be used to correctly identify calves with clinical disease, specifically diarrhea. The<br />

hypotheses for this study were that the five behavioral measures (resting position, standing<br />

posture, vigilance, lethargy and human approach test) that were selected based on sickness-<br />

driven changes in motivation would be 1) greater in younger calves when the incidence <strong>of</strong><br />

disease is highest 2) associated with variations in average daily gain (ADG), 3) more beneficial<br />

for disease identification at different ages based on motivational changes that occur as calves age<br />

and the responses to the behavioral measures will be different in calves with active diarrhea<br />

relative to healthy controls.<br />

2.2 MATERIALS & METHODS<br />

2.2.1 Animals and Housing<br />

Between October 2007 and June 2008 744 Holstein heifers housed in individual nursery<br />

pens at a commercial heifer raising facility in Western New York were enrolled in the study.<br />

Calves were born on 1 <strong>of</strong> 5 source farms located in New York State. Upon arrival at the heifer<br />

raising facility, each calf was processed according to a standard farm protocol, which included<br />

collection <strong>of</strong> a blood sample for measurement <strong>of</strong> total protein as a surrogate measure <strong>of</strong><br />

successful passive transfer <strong>of</strong> immunoglobulins from colostrum. Calves were housed<br />

individually in naturally ventilated nursery barns. Calves housed in the same nursery barn<br />

simultaneously were considered to be one enrolment group. A nursery barn consisted <strong>of</strong> 4 rows<br />

<strong>of</strong> pens with 12 pens in each row. Calves were separated by solid partitions between pens to<br />

restrict physical contact. Visual contact with calves in the adjacent row <strong>of</strong> pens was available to<br />

all calves. Each pen measured 1.2m by 2.4m. All calves were manually fed 4L <strong>of</strong> milk replacer<br />

(Excel Calf Milk Replacer 26/18, Grober Nutrition, Cambridge, Ontario, Canada) twice daily<br />

35


and had access to ad libitum calf starter (23% protein; 5% fat) and water. Weaning from milk<br />

replacer was initiated at 5 wks and completed at 6 wks <strong>of</strong> age. Calves were removed from the<br />

trial at 8 weeks <strong>of</strong> age, at which time they moved to group housing.<br />

All calves were weighed on an electronic scale (Tru-test - ID3000, Tru-Test Incorporated,<br />

Mineral Wells Texas) upon arrival at the farm (study enrollment) and at exit from the nursery<br />

facilities 8 weeks later (study completion). Calves arrived at the heifer farm at 3 ± 2 days <strong>of</strong> age<br />

(mean ± standard deviation) and weighed 41 ± 4 kg. On average, calves gained 0.61 ± 0.13 kg/d<br />

over the study period. At the end <strong>of</strong> the study, calves weighed on average 75 ± 11 kg and were<br />

59 ± 7 days old.<br />

All calves in this study were additionally enrolled in a study examining the impact <strong>of</strong> an<br />

antibiotic administered at upon arrival at the heifer raising facility on the incidence <strong>of</strong> neonatal<br />

calf diseases, specifically otitis media. Calves were injected subcutaneously with either 1.0 mL<br />

<strong>of</strong> tulathromycin (Draxxin®, Pfizer, Animal Health Group, New York, New York, USA) or a<br />

placebo at arrival. Draxxin® is a semi-synthetic macrolide antibiotic, containing 100 mg <strong>of</strong><br />

tulathromycin per mL. The placebo was a visually identical mix <strong>of</strong> propylene glycol and sterile<br />

saline.<br />

2.2.2 Behavioral Measurements<br />

Behavioral observations were performed every other week by the principle investigator,<br />

with each calf observed during three time periods (Figure 2.1). Calves were 1- 23 (mean ± SD 10<br />

± 5), 12- 37 (mean ± SD 24 ± 5), and 26- 50 (mean ± SD 38 ± 5) days <strong>of</strong> age at observation<br />

periods 1, 2, and 3, respectively.<br />

36


The observer recorded all behaviors for each calf starting with the youngest calf in the<br />

group and ending with the oldest. The observer was blind to disease diagnoses. For each calf, the<br />

observer stood immediately outside <strong>of</strong> the individual pen at the front corner, out <strong>of</strong> sight <strong>of</strong> the<br />

next calf to be observed. Milk feedings occurred at 5am and 5pm each day (Figure 2). Behavioral<br />

observations were conducted in the late morning between 10:00 and 12:00 when calves were<br />

unlikely to be anticipating their next meal. After behavioral measures were complete for an<br />

individual calf, the observer would record if calves were observed with any <strong>of</strong> the following<br />

clinical disease indicators: diarrhea, sunken eyes which indicate dehydration, elevated respiratory<br />

effort, coughing, and discharge from the nose or eyes.<br />

Behaviors <strong>of</strong> interest consisted <strong>of</strong> standing and lying postures associated with differing<br />

levels <strong>of</strong> comfort, and responses to three tests to measure lethargy relative to competing<br />

motivations <strong>of</strong> vigilance, fear, and exploration (Table 2.1). Behavioral scoring criteria were<br />

selected based on behaviors that have been previously associated with pain, discomfort, or tests<br />

such as the human approach test that has previously been used as an indicator <strong>of</strong> fear, which in<br />

dairy calves may be an indicator <strong>of</strong> malaise.<br />

Lying posture was included as a potential indicator <strong>of</strong> malaise (Hart 1988). The lying<br />

postures were evaluated on a visual analogue scale previously validated as a method <strong>of</strong> assessing<br />

acute pain in young lambs (Molony et al. 2002). It was expected that sick animals would<br />

conserve energy by restricting their exposed surface area in a short lying position. Alternatively,<br />

if there were issues with breathing due to respiratory disease, calves may lie with the neck<br />

extended in order to maximize oxygen transfer. Lateral lying was included as a posture which<br />

37


may indicate abdominal discomfort, due to either diarrhea or umbilical infection, limiting a calf’s<br />

willingness to adopt a ventral lying or short lying posture.<br />

To evaluate the value <strong>of</strong> standing posture as an indicator <strong>of</strong> illness three potential<br />

postures were identified; normal relaxed posture, statue standing, and standing with head below<br />

chest. A normal relaxed posture was expected in the majority <strong>of</strong> calves and would be evidence <strong>of</strong><br />

health. Statue standing has been observed in lambs following castration and is thought to be an<br />

indicator <strong>of</strong> visceral pain (Kent et al. 2001). Lowered head has anecdotally been observed as an<br />

indicator <strong>of</strong> illness in calves, and may represent malaise and depression during illness.<br />

An initial response test, or “Vigilance Test”, was created to determine each calf’s initial<br />

response to the observer. Four types <strong>of</strong> responses were observed; face, pr<strong>of</strong>ile, escape and<br />

ignore. It was assumed that facing the observer (face) or standing in pr<strong>of</strong>ile to the observer<br />

(pr<strong>of</strong>ile) were the normal responses to the presence <strong>of</strong> an observer. Escape was included as a<br />

measure <strong>of</strong> active withdrawal from the observer. The ignore response was expected to be the<br />

lowest vigilance response.<br />

As a measure <strong>of</strong> lethargy, the amount <strong>of</strong> observer encouragement required for lying<br />

calves to rise was measured on a 6 point scale, based on the assumption that calves would<br />

respond to observer vocalizations and invasion <strong>of</strong> their space by rising. Lower response to these<br />

signals was hypothesized to indicate that calves are conserving energy and have a strong<br />

motivation to remain lying due to exhaustion, discomfort or disinterest, which indicates illness.<br />

Due to the low number <strong>of</strong> calves that required entry into the pen (score ≥ 3) to rise during<br />

behavioral scoring at times 2 and 3, these categories were collapsed into a single combined score<br />

category <strong>of</strong> 3+.<br />

38


The human approach test used in this study is based on the knowledge that these milk-fed<br />

calves rely on humans for feed and as such will likely have positive associations with a<br />

stationary person. For this reason, it is highly probable that calves will rapidly approach a<br />

stationary observer. A previous study showed that at 17 days <strong>of</strong> age calves that were hand reared<br />

without additional positive human contact approached within 1 meter <strong>of</strong> an observer in an<br />

average <strong>of</strong> 12 seconds (Jago et al. 1999). It was hypothesized that failure to approach an observer<br />

by calves in this age group would be an indicator <strong>of</strong> a decreased interest in the environment,<br />

which has previously been associated with sickness behavior (Hart 1988; Dantzer & Kelley,<br />

2007).<br />

2.2.3 Disease Recordings and Weight Gain<br />

All disease treatment events were recorded daily by barn staff. The standard farm practice<br />

was to first record all treatments on standardized paper forms and later enter the data into a farm<br />

management computer program (DairyComp 305, Valley Agriculture S<strong>of</strong>tware Inc., Tulare,<br />

CA). Standardized definitions were used for the most common diseases. Neonatal diarrhea<br />

complex was defined as “feces s<strong>of</strong>t and does not hold form”. Non-specific fever was defined as<br />

“dull and listless with a rectal temperature above 39.5 o C”, and was distinct from the definition<br />

for respiratory disease, which was “rectal temperature above 39.5 o C, together with elevated<br />

respiratory rate, nasal discharge or cough”. Neonatal calf diarrhea complex was the most<br />

commonly treated disease with 84% <strong>of</strong> calves treated at a median age <strong>of</strong> 10 days. Only 4%<br />

(27/788) <strong>of</strong> calves were never treated for disease. Failure <strong>of</strong> passive transfer was determined<br />

based on serum total protein < 5.4 g/dL (Tyler et al. 1998). Failure <strong>of</strong> passive transfer was<br />

39


identified in 5% (40/788) <strong>of</strong> calves. Mortality was 2% (n = 14) during this study. The full<br />

description <strong>of</strong> morbidity and mortality has been published elsewhere (Chapter 5).<br />

Definition <strong>of</strong> Active Diarrhea<br />

Diarrhea was considered to be active if calves were observed by the principal investigator<br />

between 1 day pre- and 3 days post- diagnosis <strong>of</strong> diarrhea. This was based on the assumption that<br />

the farm staff identified clinical disease quickly, and recovery from diarrhea would take a<br />

minimum <strong>of</strong> 3 days.<br />

The behaviours observed in calves with active diarrhea were compared to the behaviours<br />

performed by healthy controls. Healthy controls were restricted to calves that were not identified<br />

with a clinically apparent disease by barn staff within a specific time period before and after<br />

behavioral measures were recorded by the observer (Table 2.2). A longer restriction period was<br />

applied to non-specific fever and respiratory disease based on the increased difficulty <strong>of</strong><br />

detecting these diseases. All calves with non-specific lameness were eliminated as potential<br />

controls based on the highly chronic nature <strong>of</strong> this condition in this population <strong>of</strong> calves. In<br />

addition, on the day <strong>of</strong> observation, calves observed with any <strong>of</strong> the following signs <strong>of</strong> illness<br />

were excluded as control calves; loose-watery stool, dehydration (eyes sunken), discharge from<br />

the eyes or nose, or elevated respiratory effort.<br />

2.2.4 Statistical Analysis<br />

All statistical analyses were performed using SAS® 9.1 (SAS Institute, Cary, N.C., USA)<br />

All frequencies, medians and means were generated using Proc Freq and Proc Means.<br />

40


Calves that were not observed at all three observation points, or were not weighed upon<br />

exit from the study, were excluded from these analyses (n = 52). This was done to allow for<br />

comparison <strong>of</strong> the same populations over time.<br />

Changes in proportion <strong>of</strong> calves performing behaviors over time<br />

The probabilities <strong>of</strong> a behavior (for example: short lying, ventral lying, human approach<br />

test) occurring (n = 692 calves at 3 time points) over time were analyzed using generalized linear<br />

mixed models with a logit transformation (Proc Glimmix). Calf id was included as a random<br />

effect and observation period as a fixed effect. Significance was based on a P-value ≤ 0.05.<br />

Associations <strong>of</strong> behavioral measures with average daily gain at different ages<br />

The associations between behavior and ADG were evaluated using generalized linear<br />

mixed models (Proc Mixed) with ADG as the outcome and random effects for source farm and<br />

enrolment group forced into each model. Each behavioral measurement was analyzed separately<br />

to determine the utility <strong>of</strong> specific behavioral measurements for predicting ADG and eliminate<br />

any correlation between behavioral measurements. Antimicrobial treatment, age at scoring and<br />

weight at study enrolment were included as fixed effects in each model to control for the<br />

variation in age at each observation and any effect antimicrobial treatment may have had on<br />

disease severity or growth. Behavioral scores at time periods 1, 2, and 3 were included in<br />

separate models to evaluate the impacts <strong>of</strong> behavior on ADG at different ages. Significance was<br />

based on a P-value ≤ 0.05.<br />

Associations <strong>of</strong> behaviour with neonatal calf diarrhea complex<br />

Lying position, vigilance test, lethargy, standing posture and human approach test were<br />

assessed for their associations with diarrhea. Diarrhea was <strong>of</strong> interest due to the high prevalence<br />

41


<strong>of</strong> diarrhea in calves on this farm and the importance <strong>of</strong> this disease in calves in North America.<br />

The majority <strong>of</strong> diarrhea events were observed at the first observation time, so analysis was<br />

restricted to this observation time point to control for the effect <strong>of</strong> age on behavior.<br />

Separate chi-square tests were performed to determine if there were differences in the<br />

response <strong>of</strong> calves with diarrhea versus control within each response category within the separate<br />

behavioral measures. If there were fewer than 6 observations in a response category a Fisher’s<br />

exact test was used. Significance was based on a P-value ≤ 0.05 and tendencies were discussed if<br />

the P-value was ≤ 0.10 and > 0.05.<br />

2.3 RESULTS<br />

2.3.1 Changes in response to behavior measures over time and associations with<br />

Average Daily Gain<br />

The proportion <strong>of</strong> calves that were ventral lying and short lying decreased over time (P <<br />

0.001, Table 2.3). No calves were observed short lying, lying with neck extended, lateral lying<br />

statue standing or standing with a lowered neck at all observation periods.<br />

The ADG <strong>of</strong> calves that were observed statue standing at observation 1 was 0.11 ± 0.03<br />

kg lower than calves that were observed with a normal posture (P = 0.002; Table 2.4). Calves<br />

were most likely to ignore the observer at observation 1 (P < 0.001). No calves were observed<br />

escaping or ignoring the observer at all three observation periods. Less than 1% <strong>of</strong> calves (3/692)<br />

were observed in pr<strong>of</strong>ile at all three observation periods. There was no difference in the ADG <strong>of</strong><br />

calves with different responses to the vigilance test at any <strong>of</strong> the 3 observation times.<br />

42


For the human approach test, 11% (75/692) <strong>of</strong> calves never approached the observer,<br />

28% (194/692) approached the observer during only one observation period, 33% (230/692)<br />

approached the observer at two <strong>of</strong> the observation periods, and 28%(193/692) <strong>of</strong> calves<br />

approached the observer during all three observation periods.<br />

Over the three observation periods, 42% (289/692) <strong>of</strong> calves never had a score <strong>of</strong> 3+ on<br />

the lethargy scale, 45% (309/692) <strong>of</strong> calves had a score <strong>of</strong> 3+ at one observation, 11% (78/692)<br />

<strong>of</strong> calves had a score <strong>of</strong> 3+ at two observations and 2% (16/692) <strong>of</strong> calves always had a score <strong>of</strong><br />

3+. Calves were 4.2 (95% CI: 3.3-5.3) times more likely to have a score <strong>of</strong> 0 on the lethargy<br />

scale at observation 3 relative to observation 1 (P < 0.001). The proportion <strong>of</strong> calves in the<br />

different standing postures did not change over time. Calves were 1.6 (CI: 1.3-2.0) times more<br />

likely to approach the observer at observation 3 compared to observation 1 (P < 0.001).<br />

2.3.2 Behaviors associated with neonatal calf diarrhea complex<br />

The average age <strong>of</strong> calves with diarrhea and control calves was 10 ± 4 (mean ± standard<br />

deviation) days and 9 ± 5 days, respectively. Calves with diarrhea were more likely to be short<br />

lying than their healthy cohorts (P = 0.04). Ventral lying and short lying were observed in 55%<br />

(127/229) and 12% (28/229) <strong>of</strong> diarrhea calves, respectively. Ventral lying was observed in 45%<br />

(21/47) <strong>of</strong> calves in the control group. There was no difference in the proportion <strong>of</strong> calves ventral<br />

lying in the diarrhea and control groups (P = 0.17). Lateral lying and lying with neck extended<br />

was observed in 3 calves and 1 calf in the diarrhea group, respectively. One control calf was<br />

observed short lying and no control calves were observed lateral lying or lying with neck<br />

extended. The control calf that was observed short lying was also observed shivering.<br />

43


The most common response to the vigilance test was facing the observer regardless <strong>of</strong><br />

diarrhea status (Table 2.5).<br />

A greater proportion <strong>of</strong> diarrhea calves had a lethargy score <strong>of</strong> 3+ compared to the<br />

control group (P = 0.0004; Figure 2.3). A smaller proportion <strong>of</strong> calves with diarrhea had a<br />

lethargy score <strong>of</strong> 0 compared to the healthy cohort (P = 0.0008).<br />

The control calf which exhibited statue standing was 2 days old and was shivering at the<br />

time <strong>of</strong> observation. There were too few calves observed performing either statue standing or<br />

lowered neck to perform statistical analysis. Two calves did not have a standing posture recorded<br />

due to a failure to maintain a voluntary standing posture, 1 in the control group (2 days old) and<br />

1 in the diarrhea group (7 days old).<br />

The proportion <strong>of</strong> calves that approached the observer during the human approach test<br />

was 46% (105/229) and 70% (33/47) in the diarrhea and control groups, respectively (P =<br />

0.0007).<br />

2.4 DISCUSSION<br />

Five measures <strong>of</strong> behavior were assessed based on the hypothesis that animals alter their<br />

behavior during illness. The use <strong>of</strong> different lying and standing postures have recently been used<br />

in the identification <strong>of</strong> pain associated with illness (Todd 2007) and have previously been<br />

validated for use in lambs following castration (Molony et al. 2002). The use <strong>of</strong> vigilance tests,<br />

lethargy scale and human approach tests are novel tests in the field <strong>of</strong> sickness behavior.<br />

Lying posture was divided into four categories based on the hypothesis that animals<br />

during the clinical phase <strong>of</strong> disease will alter their behavior to conserve their energy or minimize<br />

44


the discomfort associated with clinical signs. It was hypothesized that short lying would be<br />

adopted to conserve energy by reducing the surface area exposed to the environment and<br />

reducing the lower critical temperature (Schrama et al. 1993). The hypothesis was supported by<br />

the finding that 12% <strong>of</strong> animals with active diarrhea were observed short lying and only 1<br />

healthy calf (2%) was observed short lying. Due to the lower amount <strong>of</strong> disease in older calves<br />

we cannot confirm if this difference in the proportion <strong>of</strong> calves short lying was due to the lower<br />

prevalence <strong>of</strong> disease in older animals or if this behavior was restricted to neonatal calves which<br />

are the most vulnerable to temperature changes. The lower critical temperature, defined as the<br />

environmental temperature at which calves are required to exert energy to maintain their body<br />

temperature, for healthy neonates is estimated at 10.8 o C and 8.2 o C for 10 and 20 day old calves,<br />

respectively (Gonzalez-Jimenez & Blaxter, 1962). This lower critical temperature will be higher<br />

in animals with either a fever or anorexia. For calves with diarrhea the malabsorption that occurs<br />

during this disease complex would mimic fasting, which raises the lower critical temperature <strong>of</strong><br />

animals (Blaxter, 1962). Short lying was not associated with ADG at any <strong>of</strong> the three time<br />

points. The lower range in critical temperature for older calves may decrease the need <strong>of</strong> these<br />

animals to perform short lying behavior in order to conserve energy. Calves may also be altering<br />

their lying behavior during the first observation period in response to discomfort from diarrhea,<br />

since short lying was more common in the diarrhea calves.<br />

Calves resting with an extended neck were observed very rarely in the sub-sample <strong>of</strong><br />

clinically diseased calves and their healthy cohorts. For all calves in the study, the probability <strong>of</strong><br />

lying with neck extended did not change over time and was not associated with changes in ADG<br />

45


at any observation period. For this reason, the hypothesis that lying with neck extended was<br />

associated with poor oxygen exchange cannot be confirmed or denied by this study.<br />

Lateral lying was also observed very rarely, and as such no conclusions can be drawn<br />

with respect to the usefulness <strong>of</strong> this behavioral response as a tool for identifying sick animals.<br />

However, it is worth noting that no control calves in the diarrhea analysis were observed lateral<br />

lying and the calves that were lateral lying at observation 2 and 3 had very low ADG, so future<br />

study <strong>of</strong> this behavioral response may be warranted.<br />

An overall challenge when measuring lying posture in calves was the inability to observe<br />

all <strong>of</strong> the calves while they were lying down. In this study, the proportion <strong>of</strong> calves that were not<br />

observed lying was 4.0 (CI: 3.2-5.0) times greater at time period 3 compared to time period 1.<br />

This pattern is consistent with the findings <strong>of</strong> Panivivat et al (2004), who found that the lying<br />

time <strong>of</strong> calves decreased from 81% to 73% from week 1 to week 6 <strong>of</strong> life (Panivivat et al., 2004).<br />

As expected, a larger number <strong>of</strong> animals in this study were not lying at the time <strong>of</strong> observation,<br />

given that this was a single measurement taken during the day, when calves were most likely to<br />

be active. However, this absence <strong>of</strong> lying could be beneficial for identifying ill calves because<br />

69% <strong>of</strong> the calves with diarrhea were observed lying and only 47% <strong>of</strong> healthy animals were<br />

observed lying.<br />

The proportion <strong>of</strong> animals in the different standing postures did not change over time.<br />

Statue standing in period 1 was associated with decreased ADG. If statue standing is a response<br />

to clinical disease that has little impact on ADG, for example bloat or hernia which may cause<br />

discomfort, the association between statue standing and ADG will be weakened. Statue standing<br />

was only performed in calves with diarrhea, and in one calf that was not treated for disease but<br />

46


was observed shivering at the time <strong>of</strong> observation. While interpretation <strong>of</strong> these results is limited<br />

due to a small sample size, it does suggest this is an indicator <strong>of</strong> diarrhea and given the<br />

association <strong>of</strong> statue standing with pain in lambs (Molony et al. 1993) supports the hypothesis<br />

that diarrhea is painful in calves.<br />

The vigilance test used in this study had four possible responses based on observations<br />

made prior to study initiation. The proportion <strong>of</strong> calves that faced the observer increased over<br />

time such that calves were 2.7 (CI: 1.9-3.7) more likely to face the observer at observation 3<br />

relative to observation 1. It was expected that due to the lethargy and depression that has been<br />

observed as part <strong>of</strong> sickness behavior, sick calves would be more likely to initially ignore the<br />

presence <strong>of</strong> an observer (Hart 1988). Ignoring the observer and holding the head in pr<strong>of</strong>ile to the<br />

observer were less likely at time point 3 relative to time point 1. However, these were not<br />

associated with ADG using this schedule <strong>of</strong> observations. It was hypothesized that calves that<br />

tried to escape the observer may be more flighty in temperament. Studies <strong>of</strong> beef cattle have<br />

shown that flight distance is a good measure <strong>of</strong> fearfulness and is associated with lower ADG<br />

(Müller & von Keyserlingk. 2006). In this study, attempting to escape the observer was not<br />

associated with lower ADG. This may indicate that flight is not an appropriate measurement <strong>of</strong><br />

temperament for dairy calves or the temperament <strong>of</strong> dairy calves is not associated with<br />

differences in ADG <strong>of</strong> this age group. This may be due to decreased fearfulness <strong>of</strong> dairy calves<br />

due to the routine handling and the direct provision <strong>of</strong> food to the calves by people.<br />

A lethargy score <strong>of</strong> 3+ was observed in 58% and 30% <strong>of</strong> calves with diarrhea and the<br />

healthy cohort, respectively. This indicates that during active diarrhea calves require a stronger<br />

stimulus to rise and this may be useful for identifying sick animals. A score <strong>of</strong> 3+ at observation<br />

47


3 was associated with lower ADG relative to a score <strong>of</strong> 0. Although this finding was not<br />

significant at observation 1 or 2, this may indicate that this behavioral measurement is more<br />

reliably associated with reduced ADG in older animals. Only 9% <strong>of</strong> calves had a lethargy score<br />

<strong>of</strong> 3+ at observation 3 versus 49% at observation 1. As animals mature they spend less time<br />

resting and this may result in fewer false positives for this behavioral measure as an indicator <strong>of</strong><br />

illness. There are some limitations for the direct transfer <strong>of</strong> this measure to other populations<br />

because the measure itself is specific to the design <strong>of</strong> this stall. However, based on the stall<br />

dimensions and the difficulty required to enter the pen, similar approaches may be used in other<br />

facilities. Where specific cutpoints <strong>of</strong> this test are designated in other facilities will be<br />

determined by the physical barriers to the calves (i.e., gates and fences) which are <strong>of</strong>ten noisy to<br />

open and may provide the largest inducement for calves to rise. Flight zones and willingness to<br />

approach handlers can vary depending on the amount <strong>of</strong> positive or negative interactions with the<br />

animals (de Passille et al. 1996; Breuer et al. 2003). However, a score <strong>of</strong> 3+ in this study<br />

represented a flight zone <strong>of</strong> less than 1 meter with multiple vocalizations by the observer. The<br />

fact that the proportion <strong>of</strong> animals that had a score <strong>of</strong> 3+ decreased from 49% to 9% suggests that<br />

the flight zone <strong>of</strong> neonates is smaller than that <strong>of</strong> older animals. It has previously been shown<br />

that calves at two weeks <strong>of</strong> age show less exploratory behavior and are less fearful in response to<br />

a startle test than 6 week old calves (Lauber et al. 2006). The behavioral responses observed are<br />

consistent with the finding that calves required less encouragement to rise at older ages. The<br />

lethargy test would be related to the startle response as calves are encouraged by vocalization<br />

and movement to rise. At younger ages calves are less fearful <strong>of</strong> sudden movement and as such<br />

are less likely to rise and as such unhealthy calves would not stand out in the lethargy test. This<br />

48


was supported by the lack <strong>of</strong> association between a lethargy score <strong>of</strong> 3+ at observation periods 1<br />

and 2 and ADG.<br />

The proportion <strong>of</strong> calves willing to approach the observer was 70% in the healthy cohort<br />

and 44% in the diarrhea cohort. Calves with diarrhea may be in discomfort from damage to the<br />

intestinal tract through destruction and sloughing <strong>of</strong> enterocytes, villous atrophy, and<br />

inflammation <strong>of</strong> the submucosa, which may cause the greatest reluctance to approach.<br />

Alternatively, calves with diarrhea may be anorexic and lethargic which would decrease their<br />

motivation to approach an observer if they associated people with the provision <strong>of</strong> feed. The<br />

probability <strong>of</strong> calves approaching the observer was 1.6 (1.3-2.0) times greater at observation 3<br />

compared to observation 1, P < 0.001. In addition, 10% <strong>of</strong> calves never approached the observer<br />

and had the lowest ADG. This may indicate that these calves are more timid. This would also<br />

suggest that this approach may be useful for identifying calves that are more flighty and perhaps<br />

is more valuable for determining flightiness in neonatal dairy calves than the escape behavior as<br />

a first response to observers. The calves in this study were fed twice daily by farm staff. The<br />

lower ADG <strong>of</strong> calves that did not approach the observer at observations 2 or 3 may represent<br />

calves with a higher fear level which may be associated with decreased immunity, and growth<br />

(Carroll & Forsberg, 2007). Alternatively, some <strong>of</strong> these calves may be less likely to approach an<br />

observer due to decreased appetite resulting from illness (Aubert, 1999).<br />

There was a higher than normal incidence <strong>of</strong> disease in this study population. This was<br />

due to several factors but primarily due to a focus on identification and treatment <strong>of</strong> disease, and<br />

the mixing <strong>of</strong> animals from multiple sites. This study was conducted at a commercial contract<br />

heifer raising facility, which focuses on high survival rates and maintaining targeted growth rates<br />

49


<strong>of</strong> heifers to allow them to calve at 24 months <strong>of</strong> age. For this reason, early detection <strong>of</strong> disease<br />

was <strong>of</strong> extreme importance to this facility manager and this has resulted in a mortality rate <strong>of</strong><br />

1.8% in pre-weaned heifers, which is much lower compared to the average mortality rate in the<br />

United States <strong>of</strong> 7.8% (USDA 2010). For this reason, the treatment criteria used on this farm<br />

were very sensitive. For example, on this farm diagnosis <strong>of</strong> diarrhea was based on a single day <strong>of</strong><br />

diarrhea. Many farmers may not consider a single day <strong>of</strong> diarrhea to be clinical disease, and as<br />

such this would not be recorded. Additionally, calves at this facility were mixed from multiple<br />

sites, which is a known risk factor for disease (Maunsell & Donovan, 2008). As well, there was a<br />

large number <strong>of</strong> susceptible animals <strong>of</strong> the same age, which would allow pathogens to spread<br />

quickly. This high rate <strong>of</strong> disease in this population should not affect the outcomes <strong>of</strong> this study<br />

because individual calves should alter their behavior in a similar manner to disease regardless <strong>of</strong><br />

the overall incidence <strong>of</strong> disease in the population. However, these findings should be restricted to<br />

this age group <strong>of</strong> calves due to the changes in responsiveness to novelty and fearful situations<br />

with age, as well as the decreasing thermoregulatory demands <strong>of</strong> older animals. These factors<br />

will alter behavioral responses, such as short lying. Lethargy should be studied further in older<br />

animals to determine if there is an age at which fearfulness outweighs the motivation <strong>of</strong> calves<br />

during clinical disease to alter their behavior in response to illness or discomfort. If this does not<br />

occur, this would be a valuable indicator for the detection <strong>of</strong> disease in older animals.<br />

The observation schedule during this study was three short observation times over the<br />

course <strong>of</strong> approximately 8 weeks. This schedule allowed for the observation <strong>of</strong> a large number <strong>of</strong><br />

sick animals. However, there were limitations in this study design in that the health status and<br />

calves response to their internal and external motivators can change rapidly and this observation<br />

50


schedule may have resulted in these changes not being captured. Average daily gain is an<br />

unbiased measure <strong>of</strong> poor growth and by extension disease, however some illnesses may either<br />

not be associated with decreased ADG or calves may experience compensatory growth prior to<br />

their <strong>final</strong> measurement. In these cases, behavioral changes associated with these illnesses would<br />

not be detected in this study. However, given these limitations, the behaviors identified would be<br />

associated with the most severe illness which was most likely to cause production loss and<br />

impair calf welfare.<br />

In conclusion, short lying appears to be a highly specific indicator <strong>of</strong> diarrhea since it was<br />

more likely to be performed during active diarrhea. However, short lying was uncommon. For<br />

this reason, behavioral measures such as lethargy score and willingness to approach the observer<br />

may be useful to identify animals with decreased ADG and diarrhea. These behaviors could be<br />

integrated into future studies <strong>of</strong> predictors <strong>of</strong> disease and discomfort as well as used on farm for<br />

training <strong>of</strong> personnel in disease detection.<br />

51


2.5 REFERENCES<br />

Al-Chaer, E. D. & Traub, R. J. 2002. Biological basis <strong>of</strong> visceral pain: recent developments.<br />

Pain, 96, 221-225.<br />

Aubert, A. 1999. Sickness and behaviour in animals: a motivational perspective. Neuroscience<br />

& Biobehavioral Reviews, 23, 1029-1036.<br />

Blaxter, K. L. 1962. The Energy Metabolism <strong>of</strong> Ruminants. : London: Hutchinson.<br />

Borderas, T. F., de Passille, A. M. B. & Rushen, J. 2008. Behavior <strong>of</strong> dairy calves following a<br />

low dose <strong>of</strong> bacterial endotoxin. J.Anim Sci., .<br />

Borderas, T. F., Rushen, J., von Keyserlingk, M. A. G. & de Passille, A. M. B. 2009.<br />

Automated measurement <strong>of</strong> changes in feeding behavior <strong>of</strong> milk-fed calves associated with<br />

illness. Journal <strong>of</strong> dairy science, 92, 4549-4554. doi: 10.3168/jds.2009-2109.<br />

Breuer, K., Hemsworth, P. H. & Coleman, G. J. 2003. The effect <strong>of</strong> positive or negative<br />

handling on the behavioural and physiological responses <strong>of</strong> nonlactating heifers. Applied Animal<br />

Behaviour Science, 84, 3-22.<br />

Buhman, M. J., Perino, L. J., Galyean, M. L., Wittum, T. E., Montgomery, T. H. &<br />

Swingle, R. S. 2000. Association between changes in eating and drinking behaviors and<br />

respiratory tract disease in newly arrived calves at a feedlot. American Journal <strong>of</strong> Veterinary<br />

Research, 61, 1163-1168.<br />

Carroll, J. & Forsberg, N. E. 2007. Influence <strong>of</strong> stress and nutrition on cattle immunity. The<br />

veterinary clinics <strong>of</strong> North America. Food animal practice, 23, 105-149.<br />

Dantzer, R. 2001. Cytokine-Induced Sickness Behavior: Where Do We Stand? Brain, behavior,<br />

and immunity, 15, 7-24.<br />

Dantzer, R. & Kelley, K. W. 2007. Twenty years <strong>of</strong> research on cytokine-induced sickness<br />

behavior. Brain, Behavior, and Immunity, 21, 153-160.<br />

de la Fuente, R., Luzón, M., Ruiz-Santa-Quiteria, J. A., García, A., Cid, D., Orden, J. A.,<br />

García, S., Sanz, R. & Gómez-Bautista, M. 1999. Cryptosporidium and concurrent infections<br />

with other major enterophatogens in 1 to 30-day-old diarrheic dairy calves in central Spain.<br />

Veterinary parasitology, 80, 179-185.<br />

de Passille, A. M., Rushen, J., Ladewig, J. & Petherick, C. 1996. Dairy calves' discrimination<br />

<strong>of</strong> people based on previous handling. Journal <strong>of</strong> animal science, 74, 969-974.<br />

52


Goldhawk, C., Chapinal, N., Veira, D. M., Weary, D. M. & von Keyserlingk, M. A. G. 2009.<br />

Prepartum feeding behavior is an early indicator <strong>of</strong> subclinical ketosis. Journal <strong>of</strong> dairy science,<br />

92, 4971-4977.<br />

Gonzalez-Jimenez, E. & Blaxter, K. L. 1962. The metabolism and thermal regulation <strong>of</strong> calves<br />

in the first month <strong>of</strong> life. British Journal <strong>of</strong> Nutrition, 16, 199-212.<br />

Hart, B. L. 1988. Biological basis <strong>of</strong> the behavior <strong>of</strong> sick animals. Neuroscience and<br />

biobehavioral reviews, 12, 123.<br />

Huzzey, J. M., Veira, D. M., Weary, D. M. & von Keyserlingk, M. A. G. 2007. Prepartum<br />

Behavior and Dry Matter Intake Identify Dairy Cows at Risk for Metritis. Journal <strong>of</strong> dairy<br />

science, 90, 3220-3233..<br />

Jago, J. G., Krohn, C. C. & Matthews, L. R. 1999. The influence <strong>of</strong> feeding and handling on<br />

the development <strong>of</strong> the human–animal interactions in young cattle. Applied Animal Behaviour<br />

Science, 62, 137-151.<br />

Kent, J. E., Molony, V. & Graham, M. J. 2001. The Effect <strong>of</strong> Different Bloodless Castrators<br />

and Different Tail Docking Methods on the Responses <strong>of</strong> Lambs to the Combined Burdizzo<br />

Rubber Ring Method <strong>of</strong> Castration. The Veterinary Journal, 162, 250-254.<br />

Lauber, M. C. Y., Hemsworth, P. H. & Barnett, J. L. 2006. The effects <strong>of</strong> age and experience<br />

on behavioural development in dairy calves. Applied Animal Behaviour Science, 99, 41-52.<br />

Maunsell, F. & Donovan, G. A. 2008. Biosecurity and Risk Management for Dairy<br />

Replacements. Veterinary Clinics <strong>of</strong> North America: Food Animal Practice, 24, 155-190.<br />

Millman, S. T. 2007. Sickness behavior and its relevance to animal welfare assessment at the<br />

group level. Animal Welfare, 16, 123-125.<br />

Molony, V. & Kent, J. E. 1997. Assessment <strong>of</strong> acute pain in farm animals using behavioral and<br />

physiological measurements. Journal <strong>of</strong> animal science, 75, 266-272.<br />

Molony, V., Kent, J. E. & Robertson, I. S. 1993. Behavioural responses <strong>of</strong> lambs <strong>of</strong> three ages<br />

in the first three hours after three methods <strong>of</strong> castration and tail docking. Research in veterinary<br />

science, 55, 236-245.<br />

Molony, V., Kent, J. E. & McKendrick, I. J. 2002. Validation <strong>of</strong> a method for assessment <strong>of</strong><br />

an acute pain in lambs. Applied Animal Behaviour Science, 76, 215-238.<br />

Müller, R. & von Keyserlingk, M. A. G. 2006. Consistency <strong>of</strong> flight speed and its correlation<br />

to productivity and to personality in Bos Taurus beef cattle. Applied Animal Behaviour Science,<br />

99, 193-204.<br />

53


Panivivat, R., Kegley, E. B., Pennington, J. A., Kellogg, D. W. & Krumpelman, S. L. 2004.<br />

Growth Performance and Health <strong>of</strong> Dairy Calves Bedded with Different Types <strong>of</strong> Materials.<br />

Journal <strong>of</strong> dairy science, 87, 3736-3745.<br />

Proudfoot, K. L., Huzzey, J. M. & von Keyserlingk, M. A. G. 2009. The effect <strong>of</strong> dystocia on<br />

the dry matter intake and behavior <strong>of</strong> Holstein cows. Journal <strong>of</strong> dairy science, 92,.<br />

Schaefer, A. L., Cook, N. J., Church, J. S., Basarab, J., Perry, B., Miller, C. & Tong, A. K.<br />

W. 2007. The use <strong>of</strong> infrared thermography as an early indicator <strong>of</strong> bovine respiratory disease<br />

complex in calves. Research in veterinary science, 83, 376-384.<br />

Schrama, J. W., Arieli, A., Brandsma, H. A., Luiting, P. & Verstegen, M. W. 1993. Thermal<br />

requirements <strong>of</strong> young calves during standing and lying. Journal <strong>of</strong> animal science, 71, 3285-<br />

3292.<br />

Snodgrass, D., Terzolo, H., Sherwood, D., Campbell, I., Menzies, J. & Synge, B. 1986.<br />

Aetiology <strong>of</strong> diarrhoea in young calves. Veterinary Record, 119, 31-34.<br />

Svensson, C. & Jensen, M. B. 2007. Short Communication: Identification <strong>of</strong> Diseased Calves<br />

by Use <strong>of</strong> Data from Automatic Milk Feeders. J.Dairy Sci., 90, 994-997.<br />

Todd, C. G. 2007. Effects <strong>of</strong> meloxicam (metacam) therapy on the behaviour <strong>of</strong> dairy calves<br />

with neonatal calf diarrhea complex. MSc <strong>Thesis</strong> thesis, <strong>University</strong> <strong>of</strong> <strong>Guelph</strong>,.<br />

Tyler, J. W., Hancock, D. D., Wiksie, S. E., Holler, S. L., Gay, J. M. & Gay, C. C. 1998. Use<br />

<strong>of</strong> Serum Protein Concentration to Predict Mortality in Mixed-Source Dairy Replacement<br />

Heifers. Journal <strong>of</strong> Veterinary Internal Medicine, 12, 79-83.<br />

USDA. 2010. Dairy 2007, Heifer Calf Health and Management Practices on U.S. Dairy<br />

Operations, 2007. #550.0110.<br />

USDA. 2009. Dairy 2007, Part V: Changes in Dairy Cattle Health and Management Practices in<br />

the United States, 1996-2007. # 519.0709.<br />

USDA. 2007. Dairy 2007, Part I: Reference <strong>of</strong> Dairy Cattle Health and Management Practices in<br />

the United States. #N480.1007.<br />

Weary, D. M., Huzzey, J. M. & von Keyserlingk, M. A. G. 2009. Board-invited review: Using<br />

behavior to predict and identify ill health in animals. J.Anim Sci., 87, 770-777. doi:<br />

10.2527/jas.2008-1297.<br />

54


Table 2.1: Ethogram <strong>of</strong> behaviors recognized and recorded during behavioral scoring<br />

Lying Posture<br />

Ventral Lying Calf was lying with legs tucked under body or one front leg<br />

extended in front <strong>of</strong> body and head resting in relaxed and<br />

upright position<br />

Short Lying Calf was lying with head tucked tight to side <strong>of</strong> body and limbs<br />

tucked under body<br />

Neck Extended Calf was lying ventrally with neck and nose extended straight in<br />

front <strong>of</strong> body<br />

Lateral Lying Calf was recumbent with all four legs extended to the side<br />

Not lying Calf was standing at initiation <strong>of</strong> observation<br />

Standing Posture<br />

Normal Posture Calf is standing relaxed<br />

Neck Below Chest Calf is standing with head and neck below its chest<br />

Statue Standing Calf is standing still either trembling, stretching, a hunched<br />

back, a tucked in tail and tucked up abdomen (Molony &Kent.<br />

1997)<br />

Unable to stand Calf is unable to stand or maintain a standing posture following<br />

prompting by observer<br />

Vigilance Test<br />

Face Calf orients towards observer<br />

Pr<strong>of</strong>ile Calf stands with head and body parallel to observer<br />

Avoidance Calf immediately moves to the back <strong>of</strong> the pen<br />

Ignore Calf does not alter behavior in response to observer<br />

Human Approach Test<br />

Approach Calf approached observer within 30 seconds <strong>of</strong> standing<br />

No Approach Calf did not approach observer within 30 seconds <strong>of</strong> standing<br />

Lethargy Test<br />

0 Calf is standing or stands prior to any encouragement<br />

1 Calf stands in response to single vocal encouragement<br />

2 Calf stands in response after gate opens and 2nd vocal<br />

encouragement<br />

3 Calf stands in response to observer entering pen with 3rd vocal<br />

encouragement<br />

4 Observer must make contact with calf prior to calf standing<br />

5 Observer must make repeated contact with calf prior to calf<br />

standing<br />

55


Table 2.2: List <strong>of</strong> exclusions used to select healthy controls for comparison to calves with active<br />

diarrhea<br />

Disease Identified by barn staff Calves treated for disease by barn staff within<br />

the following time periods were excluded as<br />

Neonatal Calf Diarrhea Complex<br />

Bloat<br />

Navel Ill<br />

Non-specific fever<br />

Bovine respiratory disease complex<br />

Non-specific lameness<br />

healthy controls:<br />

Between 10 days prior to behavioral<br />

observation and 7 days after behavioral<br />

observation<br />

Between 10 days prior to behavioral<br />

observation and 10 days after behavioral<br />

observation<br />

Any treatment for lameness<br />

56


Table 2.3: Prevalence <strong>of</strong> behavioral responses among 692 milk-fed individually housed heifer<br />

calves at a commercial heifer raising facility at three observation periods (average age 10, 24 and<br />

38 days, respectively) with the odds <strong>of</strong> the response occurring at period 2 and 3, relative to<br />

period 1, controlling for calf as a random effect.<br />

Behavioural<br />

Measures<br />

Lying Posture<br />

Ventral Lying<br />

Short Lying<br />

Period 1 Period 2 Period 3 P-value<br />

53% (370) 40% (277) 28% (195) < 0.001<br />

9% (59) 2% (11) 1% (8) < 0.001<br />

Neck Extended < 1% (2) 1% (6) 1% (4) 0.40<br />

Lateral Lying 1% (5) < 1% (1) < 1% (1) 0.16<br />

Not lying<br />

37% (256) 57% (397) 70% (485) < 0.001<br />

Standing Posture<br />

Normal Posture 97% (671) 97% (666) 96% (665) 0.59<br />

Statue Standing 2% (14) 2% (17) 3% (20) 0.59<br />

Neck Below Chest 1% (4) 1% (6) 1% (7) 0.66<br />

Vigilance Test<br />

Face<br />

Pr<strong>of</strong>ile<br />

Escape<br />

80% (554) 91% (627) 92% (634) < 0.001<br />

11% (79) 6% (39) 6% (42) < 0.001<br />

1% (8) 3% (20) 1% (5) 0.006<br />

Ignore<br />

7% (51) 1% (6) 1% (9) < 0.001<br />

Human Approach Test<br />

Approach<br />

52% (359) 63% (436) 63% (439) < 0.001<br />

No Approach<br />

Lethargy<br />

48% (333) 37% (256) 37% (254)<br />

0<br />

41% (284) 61% (422) 74% (514) < 0.001<br />

1<br />

2<br />

3+<br />

3% (18) 6% (42) 5% (38) = 0.01<br />

8% (53) 16% (113) 11% (80) < 0.001<br />

49% (337) 17% (115) 9% (61) < 0.001<br />

57<br />

Odds Ratio<br />

(Confidence Interval)<br />

Period 2 Period 3<br />

0.6<br />

(0.5-0.7)<br />

0.2<br />

(0.1-0.3)<br />

2.3<br />

(1.9-2.9)<br />

2.4<br />

(1.8-3.3)<br />

0.5<br />

(0.3-0.7)<br />

2.6<br />

(1.1-5.8)<br />

0.1<br />

(0.1-0.3)<br />

1.6<br />

(1.3-2.0)<br />

2.3<br />

(1.8-2.8)<br />

2.4<br />

(1.4-3.9)<br />

2.4<br />

(1.7-3.3)<br />

0.2<br />

(0.2-0.3)<br />

0.3<br />

(0.3-0.4)<br />

0.1<br />

(0.1-0.3)<br />

4.0<br />

(3.2-5.0)<br />

2.7<br />

(1.9-3.7)<br />

0.5<br />

(0.3-0.7)<br />

0.6<br />

(0.2-1.9)<br />

0.2<br />

(0.1-0.3)<br />

1.6<br />

(1.3-2.0)<br />

4.2<br />

(3.3-5.3)<br />

2.2<br />

(1.2-3.9)<br />

1.6<br />

(1.1-2.3)<br />

0.1<br />

(0.1-0.1)


Table 2.4: The lsmeans ± S.E. <strong>of</strong> the average daily gain (ADG) (kg/day) <strong>of</strong> calves (n = 692) in<br />

each Behavioral Measure at the three observation periods, controlling for age at observation,<br />

antibiotic treatment and weight at arrival as fixed effects, with source farm and enrolment cohort<br />

as random effects. P-values are based on the association <strong>of</strong> the Behavioral Measure with ADG.<br />

Behavioral Measure Period 1 Period 2<br />

Period 3<br />

Response<br />

ADG (kg/day) ADG (kg/day) ADG (kg/day)<br />

Lying Posture P = 0.60 P = 0.27 P = 0.05<br />

Ventral Lying 0.62 ± 0.02 0.61 ± 0.02 0.60 ± 0.02<br />

Short Lying 0.60 ± 0.02 0.55 ± 0.04 0.54 ± 0.05<br />

Neck Extended 0.54 ± 0.09 0.60 ± 0.05 0.54 ± 0.06<br />

Lateral Lying 0.65 ± 0.06 0.30 0.36<br />

Not applicable 0.61 ± 0.02 0.62 ± 0.02 0.62 ± 0.02<br />

Standing Posture P = 0.002 P = 0.29 P = 0.14<br />

Normal Posture 0.62 ± 0.02 a 0.62 ± 0.02 0.62 ± 0.02<br />

Statue Standing 0.50 ± 0.04 b 0.57 ± 0.03 0.56 ± 0.03<br />

Neck Below Chest 0.55 ± 0.06 a,b 0.61 ± 0.05 0.62 ± 0.05<br />

Vigilance Test P = 0.83 P = 0.22 P = 0.73<br />

Face 0.61 ± 0.02 0.61 ± 0.02 a,b 0.61 ± 0.02<br />

Pr<strong>of</strong>ile 0.62 ± 0.02 0.63 ± 0.02 a 0.60 ± 0.02<br />

Escape 0.64 ± 0.05 0.61 ± 0.03 a,b 0.64 ± 0.06<br />

Ignore 0.62 ± 0.02 0.52 ± 0.05 b 0.62 ± 0.04<br />

Human Approach Test P= 0.57 P = 0.11 P = 0.003<br />

Approach 0.62 ± 0.02 0.62 ± 0.02 0.62 ± 0.02<br />

No Approach 0.61 ± 0.02 0.60 ± 0.02 0.60 ± 0.02<br />

Lethargy P = 0.99 P = 0.39 P = 0.006<br />

0 0.62 ± 0.02 0.62 ± 0.02 0.62 ± 0.02 a<br />

1 0.61 ± 0.03 0.59 ± 0.02 0.63 ± 0.02 a,b<br />

2 0.62 ± 0.02 0.60 ± 0.02 0.59 ± 0.02 b,c<br />

3+ 0.61 ± 0.02 0.61 ± 0.02 0.57 ± 0.02 c<br />

a,b,c : If lowercase superscripts are different within a behavioral measure and column then P < 0.05<br />

indicating that ADG is significantly different between the different responses <strong>of</strong> the behavioural<br />

measure<br />

58


Table 2.5: Proportional responses <strong>of</strong> diarrhea and control calves to the observer during vigilance<br />

testing<br />

Response Control Calves Diarrhea P-value<br />

Face 89% (42/47) 77% (176/229) 0.06*<br />

Pr<strong>of</strong>ile 2% (1/47) 11% (26/229) 0.06*<br />

Escape 4% (2/47) < 1% (1/229) 0.08 ψ<br />

Ignore 4% (2/47) 11% (26/229) 0.14 ψ<br />

* P-value is based on statistical results from chi-square test<br />

Ψ P-value is based on statistical results from Fisher’s exact test<br />

59


Figure 2.1: Timeline describing trial milestones and behavioral observations, with average ages<br />

(mean ± s.d.) at observation and measurements<br />

60


Figure 2.2: Method for behavioral scoring <strong>of</strong> the calves, conducted every other week<br />

61


Proportion <strong>of</strong> calves<br />

performing behavior<br />

70%<br />

60%<br />

50%<br />

40%<br />

30%<br />

20%<br />

10%<br />

0%<br />

P=0.0008<br />

P=1.00<br />

0 1 2 3+<br />

Lethargy Score<br />

Figure 2.3: Proportion <strong>of</strong> calves in each lethargy response category for the active diarrhea (n =<br />

229) and healthy controls (n = 47)<br />

62<br />

P=0.54<br />

P<br />

P=0.004<br />

Control<br />

Diarrhea


CHAPTER 3: ASSOCIATIONS AMONG ACTIVITY PATTERNS AND WEIGHT<br />

GAIN IN NEWLY WEANED DAIRY CALVES<br />

3.1 INTRODUCTION<br />

At the time <strong>of</strong> weaning dairy calves are usually subjected to a wide variety <strong>of</strong> stressors.<br />

These stressors include not only nutritional changes, through the removal <strong>of</strong> milk from the diet,<br />

but also transportation and changes in both the physical and social environment. These may<br />

include moving from individual to group housing with major changes in social interactions as<br />

well as the introduction <strong>of</strong> novel feeding and drinking apparatuses. The first few days following<br />

weaning are characterized by changes in activity patterns including a high rate <strong>of</strong> vocalizations<br />

(Jasper et al., 2008), increased walking (Haley et al., 2005) and more time spent standing<br />

(Budzynska and Weary, 2008), all <strong>of</strong> which are thought to be indicators <strong>of</strong> distress during the<br />

post-weaning transition period (Weary et al., 2008).<br />

Different approaches have been used to reduce the stress associated with this transition<br />

period. To address the nutritional stress <strong>of</strong> weaning it is recommended that calves consume at<br />

least 0.67 kg <strong>of</strong> calf starter per day prior to weaning (Davis and Drackley, 1998). However, even<br />

when calf starter consumption is above this level, calves still showed increases in vocalizations<br />

and in the amount <strong>of</strong> time standing following weaning (Budzynska and Weary, 2008). Another<br />

approach that can be used to reduce the nutritional stress associated with weaning is to provide<br />

extended access to the feeding apparatus (pail or bucket) used during the nursery period<br />

(Budzynska and Weary, 2008; Jasper et al., 2008). Following abrupt weaning both control<br />

animals and animals provided access to a milk feeding apparatus responded with increased<br />

vocalizations, and increased time standing. However, access to a milk feeding apparatus resulted<br />

63


in a lower rate <strong>of</strong> vocalizations and tended to decrease the number <strong>of</strong> standing bouts in<br />

experimental calves compared to controls, suggesting that this approach reduced the stress<br />

response <strong>of</strong> calves to weaning.<br />

Dairy calves are commonly housed individually prior to weaning and housed in groups<br />

after weaning (USDA, 2010). They are <strong>of</strong>ten transported from a nursery to another facility,<br />

which may be located at the same site or many kilometers away. Transportation <strong>of</strong> calves has<br />

been shown to be stressful and is associated with increased risk <strong>of</strong> morbidity and mortality<br />

(Swanson& Morrow-Tesch, 2001). Following transport, calves are also <strong>of</strong>ten exposed to their<br />

first social interactions with other calves. First mixing and movement <strong>of</strong> pair-housed milk-fed<br />

calves has been shown to be associated with a decrease in lying, an increase in time spent<br />

standing, walking, social aggression, and environmental exploration (Veissier et al., 2001). For<br />

newly weaned calves, the new environment may also have different feeding and drinking<br />

apparatuses than the calves were previously exposed to. This novelty could also limit their ability<br />

to access feed if calves must learn to access the new apparatuses.<br />

Although it is generally accepted that the period <strong>of</strong> weaning and movement to group<br />

housing is a critical time in the management <strong>of</strong> dairy calves, to date, little research has been<br />

conducted to determine the individual variation in the behavioural response to weaning and how<br />

this variation is associated with weight gain. The objective <strong>of</strong> this study was to investigate the<br />

association between the individual variation in post-weaning behavioral changes and post-<br />

weaning weight gain. To address this objective the following hypotheses were tested: 1) calves<br />

that have the greatest change in activity post-weaning will have the lowest weight gain; 2) calves<br />

64


that are slower to locate and consume feed will have lower weight gains; 3) calves that rest<br />

further from their conspecifics will have lower body weight gain.<br />

3.2 MATERIALS AND METHODS<br />

This study used data collected during an unpublished experimental trial conducted<br />

between October 2006 and September 2007 in which behavioral and weight gain data were<br />

collected for 74 Holstein calves from two research herds. A wide variation in weight gain was<br />

observed in this group <strong>of</strong> calves and is the basis <strong>of</strong> the current study. Calves were 56 ± 7.4 (mean<br />

± standard deviation) days <strong>of</strong> age at the time <strong>of</strong> enrolment.<br />

Enrolment (Day 0) occurred on the day that calves received their last milk meal and were<br />

moved from individual to group housing. Enrolment and movement to group housing occurred<br />

between 09:00 and 11:00 on Day 0. Calves were transported to group housing at the study site<br />

where they were weighed with an electronic scale (Chute Scale, Avery Weigh-Tronix, Fairmont,<br />

Minnesota, USA), and height was recorded using a measuring stick. At this time, individual<br />

calves were randomly assigned to receive one injection <strong>of</strong> either the antibiotic tulathromycin (2.5<br />

mg/kg subcutaneous; Draxxin ®, Pfizer Animal Health, Montreal, Canada) or a placebo as part<br />

<strong>of</strong> an experimental trial.<br />

Calves were housed at one <strong>of</strong> two research facilities prior to enrolment in this trial.<br />

Calves at Nursery 1 (n = 50) were housed in individual pens in a warm barn, maintained between<br />

7 and 33 o C, located at the same site as the trial facility, and were transported less than 0.75 km.<br />

Nursery 2 was located 7 km from the study facility. Calves at Nursery 2 (n = 24) were housed<br />

outside in individual straw-bedded polyurethane hutches. Both facilities are associated with the<br />

<strong>University</strong> <strong>of</strong> <strong>Guelph</strong> and treatment protocols and veterinary care were consistent between the<br />

65


sites. During the summer months both farms fed calves 2L <strong>of</strong> whole milk twice per day and<br />

provided ad libitum water and calf starter (Floradale Standard 20% calf starter with Bovatec;<br />

Flordale, Ontario). In the winter months, calves at Nursery 2 were fed 2L <strong>of</strong> whole milk three<br />

times per day but were not provided any water while calves at Nursery 1 were fed as during the<br />

summer.<br />

Calf starter intake was measured daily at both facilities for two days prior to weaning to<br />

generate an average pre-weaning calf starter intake. Average calf starter intake was based on a<br />

single day <strong>of</strong> calf starter intake for four calves on the trial due to data collection error. Calf<br />

starter intake was dichotomized at 0.91 kg/d in the two days prior to weaning based on the<br />

nutrient requirements for maintenance and a modest weight gain for a 59 kg calf (Davis and<br />

Drackley, 1998).<br />

Following weaning, calves were housed in groups <strong>of</strong> 4 to 6 calves in a naturally<br />

ventilated barn located at the same site as Nursery One. Group size was determined by calving<br />

patterns on the farms and the availability <strong>of</strong> animals to form groups comprised <strong>of</strong> calves that<br />

were within 3 weeks <strong>of</strong> age <strong>of</strong> each other with an average variation in age within the group <strong>of</strong> 12<br />

days. Each pen was 3.6 meters by 7.0m with a shaving bedded pack (3.6m by 4.6m) located at<br />

the back <strong>of</strong> the pen with the remaining pen consisting <strong>of</strong> a cement section which was scraped<br />

daily. The feed bunk was located along the front <strong>of</strong> the pen and measured 3.6 m long with 9<br />

stanchions. Ambient temperature in the pen was based on a single measurement taken on Day 4<br />

using a handheld digital temperature gage.<br />

Barn staff monitored calves daily. In the event <strong>of</strong> any health concerns, the herd<br />

veterinarian was consulted. Calves were re-weighed on Day 7. Change in calf body weight<br />

66


anged from a loss <strong>of</strong> 4.5 kg to a gain <strong>of</strong> 11.5 kg with a mean gain <strong>of</strong> 3.9 ± 3.4 kg (mean ±<br />

standard deviation). Three categories <strong>of</strong> weight gain in the first week after weaning were<br />

determined by the first and third quartile <strong>of</strong> weight gain for all 74 calves. This resulted in a low<br />

weight gain category (LOW) with weight change between -4.5 and 2kg, a moderate weight gain<br />

group (MOD) between 2.5 and 6.0 kg and a high weight gain group (HIGH) between 6.5 and<br />

11.5 kg <strong>of</strong> weight gain.<br />

3.2.1 Behavior Measurements<br />

An IceTag (IceRobotics, UK) was attached to the rear metacarpal <strong>of</strong> all calves on the<br />

study on Day-3 and removed on Day 14. These devices recorded steps, daily lying bouts, daily<br />

lying bout duration, and daily lying duration for each day <strong>of</strong> the trial. Data logger malfunctions<br />

occurred for 4 calves and these animals were removed from all analyses <strong>of</strong> steps, daily lying<br />

bouts, daily bout length and daily lying duration.<br />

A subset <strong>of</strong> calves was recorded using 24-hr time lapse video (Panasonic WV-CP244<br />

Color CCTV Cameras, Heads WV-LZA61/2) between Day 0 and Day 7. One camera was<br />

attached above the bunk to record bunk behavior, and 2 cameras were attached to the walls <strong>of</strong> the<br />

pens with overlapping views <strong>of</strong> the front and the back <strong>of</strong> the pen. Calves were identified using<br />

photographs <strong>of</strong> unique markings on calves. At the time <strong>of</strong> enrolment, if calves were determined<br />

to be difficult to identify (i.e. all black calves) coloured Vet Wrap (3M, St. Paul, Mn) was<br />

wrapped around the girth <strong>of</strong> the calf. The video tape was analyzed using 5 minute instantaneous<br />

scan samples between 12:00-19:00 on Day 0, and 07:00-19:00 on Days 1 and 2. Times were<br />

selected based on the completion <strong>of</strong> all animal handling in the barn on the day <strong>of</strong> weaning and<br />

the hours in which light was sufficient to view calves. Observers were blind to both antibiotic<br />

67


treatment and weight gain category. Bunk behaviors were observed for the last 65 calves on the<br />

trial, and resting behaviors were observed for the last 44 calves on the trial (Table 3.1). Smaller<br />

sample sizes were based on the availability <strong>of</strong> video tapes with earlier groups not recorded. Bunk<br />

engaged was use to determine if calves stood at the feed bunk but did not consume feed. Bunk<br />

directed behavior was used as an indicator <strong>of</strong> reluctance to place the head through the stanchion,<br />

which is potential inhibitor <strong>of</strong> feed intake. Close and Far Lying were used as measurements <strong>of</strong><br />

affiliation. One group <strong>of</strong> 5 calves was excluded from the study <strong>of</strong> bunk behaviors due to the<br />

failure <strong>of</strong> video recording equipment.<br />

3.2.2 Experimental Study Design<br />

The objective <strong>of</strong> the underlying trial was to identify behavioral and physiological changes<br />

associated with bovine respiratory disease complex (BRD) or administration <strong>of</strong> a long-acting<br />

antibiotic. No calves were diagnosed with BRD in the first week post-weaning. On days 0, 4, 7,<br />

11 and 14 <strong>of</strong> the present study calves were haltered and blood samples were collected via jugular<br />

venipuncture for measures associated with the antibiotic trial.<br />

3.2.3. Statistical Analysis<br />

All statistical analyses were conducted using Statistical Analysis S<strong>of</strong>tware (SAS version<br />

9.1; SAS Institute Inc, Cary, NC, USA).<br />

Means and standard errors for each weight gain category were generated for normally<br />

distributed continuous variables (Proc Means). If the variables were not normally distributed<br />

upon visual assessment, medians were presented with a range for the 1 st and third quartiles (Proc<br />

Means).<br />

68


From the post-hoc visual assessment <strong>of</strong> the daily activity patterns <strong>of</strong> calves, the number<br />

<strong>of</strong> steps per day (Figure 3.1), number <strong>of</strong> daily lying bouts (Figure 3.2 ) and daily lying duration<br />

(Figure 3.3 ) were divided into 3 time periods: pre-weaning (days -2 and -1), transition (days 0, 1<br />

and 2), and plateau (Days 3-11). Average activity for each <strong>of</strong> these periods was calculated,<br />

resulting in 3 variables per behaviour. All behaviors (number <strong>of</strong> steps per day, daily lying<br />

duration and daily lying bouts) were assessed separately for their association with weight gain,<br />

using Day 7 weight with Day 0 weight as a covariate, using generalized linear mixed models<br />

(Proc Mixed) with a random effect for group nested within group size. The average activity for<br />

each <strong>of</strong> the 3 time periods were included as fixed effects in the analysis for each <strong>of</strong> the following<br />

activities: number <strong>of</strong> steps per day, daily lying duration and number <strong>of</strong> daily lying bouts.<br />

Eating, bunk engaged, and bunk directed behaviour between Day 0 and Day 2 were<br />

calculated as a percentage <strong>of</strong> total time at the bunk. The association between these behaviors and<br />

Day 7 weight were analyzed in separate models using generalized linear mixed models with Day<br />

0 weight and percentage <strong>of</strong> total time spent performing all bunk behaviours as covariates, with a<br />

random effect <strong>of</strong> group nested within group size.<br />

The average percentages <strong>of</strong> time that calves performed Close and Far Lying between<br />

Day 0 and Day 2 were calculated as a percentage <strong>of</strong> total lying time. The association between<br />

these behaviors and Day 7 weight were analyzed in separate models using generalized linear<br />

mixed models with Day 0 weight and percentage <strong>of</strong> total time spent lying as covariates, with a<br />

random effect <strong>of</strong> group nested within group size.<br />

All generalized linear mixed models were built using backwards step-wise regression. All<br />

models controlled for fixed effects <strong>of</strong> Day 0 body weight, nursery location, antibiotic treatment,<br />

69


pre-weaning calf starter intake, and ambient temperature with a random effect <strong>of</strong> group nested<br />

within group size. Interactions between nursery location and behavioral measures were tested for<br />

all behavior models.<br />

For all models statistical significance was defined as a P-value ≤ 0.05. The co-efficients<br />

for each variable in the models and their associated p-values were monitored for confounding.<br />

Normality <strong>of</strong> residuals was assessed using boxplots and normal probability plots. No<br />

transformations were required. All graphs that presented results <strong>of</strong> models that included<br />

continuous variables were restricted to the interquartile range <strong>of</strong> the continuous variable.<br />

3.3 RESULTS<br />

The average calf starter intake prior to weaning was 1.21 ± 0.62 (mean ± standard<br />

deviation) kg, 1.33 ± 0.79 and 0.92 ± 0.74kg for the High, Average and Low weight gain groups,<br />

respectively. Calves from Nursery 1 consumed 0.48 ± 0.1 kg more calf starter than calves from<br />

Nursery 2 (P < 0.01). Consumption <strong>of</strong> more than 0.91 kg <strong>of</strong> calf starter in the pre-wean period<br />

was associated with a 1.7 ± 0.7 kg increase in Day 7 body weight, controlling for Day 0 weight<br />

as a fixed effect and group nested within group size as a random effect (P=0.03). Calves that<br />

received antibiotic treatment weighed 1.1 ± 0.6 kg more than calves that did not receive<br />

antibiotic treatment (P=0.06).<br />

A total <strong>of</strong> 23 (31%) calves were treated for clinical disease in the 7 to 30 days following<br />

weaning. Of those treated, 91% (21/23) were treated for coccidiosis, and 9% (2/74) were treated<br />

for BRD. Day 7 weight was not associated with the probability <strong>of</strong> being treated in the 30 days<br />

following weaning, (P=0.90).<br />

70


3.3.1 Activity level in response to weaning and relocation<br />

The average number <strong>of</strong> steps per day in the pre-weaning, transition and plateau periods<br />

were 1,366 ± 43, 3,460 ± 138, and 1,585 ± 43 steps per day, respectively. The average number <strong>of</strong><br />

steps per day during the transition and plateau periods were associated with Day 7 weight,<br />

controlling for Day 0 weight, nursery location and antibiotic treatment as fixed effects and group<br />

nested within group size as a random effect (P < 0.001 and P < 0.001; Table 3.2).<br />

The average lying duration per day during the pre-weaning, transition and plateau periods<br />

were 17.4 ± 1.1 (mean ± s.d.), 13.3 ±1.5, and 15.0 ± 1.2 hours, respectively. An interaction<br />

between nursery location and daily lying duration during the transition period was significantly<br />

associated with Day 7 weight, controlling for daily lying duration during the plateau periods and<br />

Day 0 weight as fixed effects and group nested within group size as a fixed effect, (P=0.01;<br />

Figure 3.4). This result indicates that for Nursery 1 a longer lying duration tended to be<br />

negatively associated with body weight compared to Nursery 2 calves which had a positive<br />

association between body weight and lying duration. During the plateau period, increased lying<br />

duration was associated with increased Day 7 weight (Table 3.2; Figure 3.4).<br />

The average number <strong>of</strong> lying bouts during the pre-weaning, transition and plateau periods<br />

were 19 ± 0.4, 21 ± 5.6, and 13 ± 0.5, respectively. An increase in the number <strong>of</strong> lying bouts<br />

during the plateau period was associated with increased Day 7 weight (P=0.03). The number <strong>of</strong><br />

lying bouts in the transition period was not associated with <strong>final</strong> weight (P=0.15).<br />

3.3.2 Bunk behavior<br />

Bunk behavior was examined for 65 calves with 28% (n=18/65), 46% (30/65) and 26%<br />

(17/65) <strong>of</strong> observed calves in the LOW, MOD and HIGH groups, respectively.<br />

71


Calves spent a total <strong>of</strong> 13 ± 6 (mean ± standard deviation) percent <strong>of</strong> their time<br />

interacting with the bunk (eating, bunk directed and bunk engaged) during the first three days<br />

post-weaning. There was no difference in percentage <strong>of</strong> time interacting with the bunk by<br />

nursery location (P=0.12). Calves spent 49 ± 11 and 20 ±11 percent <strong>of</strong> their time at the bunk<br />

eating and bunk engaged, respectively.<br />

An interaction between total time at the bunk and nursery location was significantly<br />

associated with Day 7 body weight, (P=0.03). More time at the bunk was associated with a<br />

lower weight gain in Day 7 in calves from Nursery 1 and a greater Day 7 weight in calves from<br />

Nursery 2, controlling for Day 0 weight, antibiotic treatment and pre-weaning calf starter intake<br />

as fixed effects and group nested within group size as random effects (Figure 3.5).<br />

There was a non-significant tendency for a 1% increase in the time spent eating to be<br />

associated with a 0.07 ± 0.04 kg increase in Day 7 weight (P=0.07), controlling for Day 0<br />

weight, antimicrobial treatment, calf starter intake, nursery location and total time at the bunk as<br />

fixed effects, and group nested within group size as random effects.<br />

A 1% increase in time in bunk engaged behaviour was associated with a 0.08 ± 0.03 kg<br />

decrease in Day 7 body weight (P = 0.02), controlling for nursery location, total time at the feed<br />

bunk, antibiotic treatment, and calf starter intake as fixed effects and group nested within group<br />

size as random effects. Bunk directed behavior was not associated with Day 7 weight (P =<br />

0.53).<br />

72


3.3.3 Resting Behavior<br />

In the subset measured, 30% (13/44), 48% (21/44) and 23% (10/44) <strong>of</strong> calves were in the<br />

LOW, MOD, and HIGH week 1 weight change categories, respectively. The percentage <strong>of</strong> total<br />

lying time as measured by video observation was highly correlated with lying duration as<br />

measured by the data loggers during the transition period, (P < 0.001, R = 0.87) which indicated<br />

a high degree <strong>of</strong> agreement between the two measurements.<br />

During the first 3 days after weaning, calves spent 49 ± 10 % (mean ± s.d.) <strong>of</strong> their time<br />

lying down. Calves from Nursery 1 and 2 were observed Close lying 37 ±10% and 25 ± 10 % <strong>of</strong><br />

their total resting time, respectively (P = 0.003). Close lying was not associated with Day 7<br />

weight, controlling for Day 0 weight and total lying time as fixed covariates and group nested<br />

within group size as random effects (P = 0.38). Calves from Nursery 1 and Nursery 2 spent 0.6<br />

(q1-q3: 0-0.9) and 5.3 (q1-q3: 0.8-4.1) percent <strong>of</strong> total lying time Far lying, respectively (P =<br />

005). A 1% increase in Far lying was associated with 1.4 ± 0.6 kg decrease in Day 7 weight,<br />

controlling for Day 0 weight and total lying time as fixed effects and group nested within group<br />

size as a random effect (P = 0.02). When Nursery Location was included in the above model as a<br />

fixed effect, Far lying was not associated with Day 7 weight (P=0.17).<br />

3.4 DISCUSSION<br />

Certain changes in activity are considered to be indicative <strong>of</strong> distress in newly weaned<br />

calves. In the current study, a higher level <strong>of</strong> activity during the first three days following<br />

weaning and movement to a novel environment was significantly associated with decreased short<br />

term weight gain. This increased activity in the transition period was also found during two-<br />

staged weaning <strong>of</strong> beef calves. During the first stage <strong>of</strong> the two-staged weaning process, when<br />

73


calves were still allowed access to the dam but were restricted from sucking, beef calves walked<br />

more than control animals that could still suckle their dams (Haley et al., 2005). In the second<br />

stage <strong>of</strong> the two-staged weaning process, when all calves were removed from the dam, two-<br />

staged weaned calves walked less than control animals. While comparison between the results <strong>of</strong><br />

the two-stage weaning study and the current study must be tempered by different social stressors<br />

(isolation versus mixing stress) it does suggest a common response to a weaning and social<br />

stress.<br />

In contrast to the transition period, calves that had higher activity between 4 and 11 days<br />

post-weaning gained more weight. This may further explain the results <strong>of</strong> the study <strong>of</strong> Chua et al<br />

(2002), in which pair housed milk fed calves were more active than individually housed calves<br />

during the 8 week study period, which included weaning. However, a positive association<br />

between growth and activity following an initial adaptation to weaning has not been documented<br />

prior to the current study.<br />

A greater lying duration was positively associated with higher body weight gain during<br />

the first week post-weaning for calves from Nursery 2 during the transition period and for all<br />

calves during the plateau period. In previous studies, time spent standing was not different for<br />

calves that were provided warm water post-weaning to alleviate weaning distress in comparison<br />

with calves that were abruptly weaned despite other behavioral indicators <strong>of</strong> decreased weaning<br />

distress (Budzynska and Weary, 2008). However, time spent resting was higher after weaning for<br />

beef and dairy calves with the two-stage weaning (Haley et al., 2005; Haley, 2006). In addition,<br />

the lying time <strong>of</strong> calves decreases following mixing with unfamiliar animals (Raussi et al.,<br />

2005). This indicates that in general, calves respond to these stressors by decreasing resting time.<br />

74


Although the number <strong>of</strong> lying bouts decreased in the transition period for all calves, the<br />

absolute number <strong>of</strong> lying bouts was not associated with rate <strong>of</strong> weight gain. Combined with the<br />

results <strong>of</strong> total lying duration, this finding indicates that the total time resting may be more<br />

important than uninterrupted resting time. This is in contrast to the report by Budzynska and<br />

Weary (2008), who found that the number <strong>of</strong> lying bouts decreased in response to changes in<br />

weaning management with no change in lying time. However, in that study, calves were<br />

individually housed whereas in the present study calves were group housed. It is likely that the<br />

group housing influenced their behavior, with the activities <strong>of</strong> conspecifics disrupting or altering<br />

their resting behavior. For this reason, it is important to take into context the environment the<br />

calves are housed in when interpreting the relationship between behavior and growth since<br />

environment likely alters the response <strong>of</strong> calves to stimuli.<br />

There was an interaction between total lying duration and nursery location during the<br />

transition period. Calves from Nursery 1 that gained less weight also spent more total time lying<br />

during the transition period. However, the difference between calves that rested 12 compared to<br />

14 hours was negligible indicating that this behavior was not a strong indicator <strong>of</strong> poor growth in<br />

this group <strong>of</strong> calves. Greater weight gain was associated with more lying time for calves from<br />

Nursery 2. However, Nursery 2 calves had lower rates <strong>of</strong> gain than calves from Nursery 1. This<br />

difference in the association between lying duration and weight gain between nursery locations<br />

indicate that facility and management factors should be further investigated. Potential reasons for<br />

this difference are the lower calf starter intake <strong>of</strong> calves pre-weaning, the limited socialization <strong>of</strong><br />

calves from Nursery 2 pre-weaning and the increased transportation distance <strong>of</strong> calves from<br />

Nursery 2.<br />

75


It was hypothesized that the more time calves spent at the bunk during the transition<br />

period, the more weight they would gain in the week after weaning. This hypothesis was<br />

supported by a positive association between increased time at the bunk and post-weaning weight<br />

gain for calves from Nursery 2. However, calves from Nursery 1 had lower post-weaning weight<br />

gain with more time at the bunk. When the different interactions with the bunk were examined, it<br />

was found that, as hypothesized, an increase in time spent eating was positively associated with<br />

post-weaning weight gain and the proportion <strong>of</strong> time spent bunk engaged was negatively<br />

associated with post-weaning weight gain. This may indicate that calves are spending time at the<br />

bunk due to social facilitation but are not consuming feed. This failure to consume feed may be<br />

inappetance due to sub-clinical illness, stress, or a combination <strong>of</strong> both.<br />

The lack <strong>of</strong> association between bunk directed behavior and post-weaning weight gain<br />

may be due to conflicting motivations for this behavior. Two potential reasons for bunk directed<br />

behavior include environmental exploratory behaviors, or unwillingness to place the head<br />

through a novel feeding apparatus. These two motivations would be indicators <strong>of</strong> opposite<br />

welfare issues in that environmental exploration is a positive welfare indicator while fear from a<br />

novel feeding apparatus hindering feed intake would indicate a negative welfare state. However,<br />

using this study design there was no way to identify which motivation was responsible for the<br />

bunk directed behavior.<br />

Close lying was not associated with Day 7 weight. However, Far lying was negatively<br />

associated with Day 7 weight, when Nursery Location was not included in the model. This<br />

negative association is consistent with the original hypothesis that calves which interact less with<br />

their conspecifics will have lower weight gain. However, Nursery Location was highly correlated<br />

76


with Far lying and as such could not be included in the model. These results were interesting in<br />

that Close lying was most prevalent in calves from Nursery 1 while Far lying was most<br />

prevalent in calves from Nursery 2. Although causation cannot be determined due to the small<br />

number <strong>of</strong> nursery locations used it did suggest an effect <strong>of</strong> rearing conditions on social<br />

interactions. Calves from Nursery 2 were raised in an environment where they had limited visual<br />

contact and no physical contact with other calves. At Nursery 1 they had visual contact with a<br />

number <strong>of</strong> calves and physical contact, limited by bars between pens, to other calves. This may<br />

have allowed calves from Nursery 1 to become more familiar with each other and have more<br />

experience interaction with conspecifics through visual or olfactory cues. This familiarity was<br />

less likely to be gained by calves from Nursery 2 given their increased isolation. This difference<br />

in prior social experience may explain the difference in the performance <strong>of</strong> Far lying. Socially<br />

isolated cattle appeared to interact less with group reared animals (Broom and Leaver, 1978) .<br />

The duration <strong>of</strong> partial social isolation in the current study was less than 3 months and <strong>of</strong> lower<br />

severity compared to the 8 months <strong>of</strong> total social isolation in the study by Broom and Leaver<br />

(1978). However, it appears that even this short period <strong>of</strong> isolation might alter the social<br />

behavior <strong>of</strong> dairy calves. The study by Broom and Leaver (1978) found no effect <strong>of</strong> isolated<br />

rearing on weight gain. However, the lower weight gains observed in calves from Nursery 2 and<br />

the higher rate <strong>of</strong> Far lying does suggest that pre-weaning social environment may have an<br />

effect on post-weaning behavior and weight gain.<br />

The calves in this study had various levels <strong>of</strong> calf starter intake with higher pre-weaning<br />

calf starter intakes associated with greater post-weaning gain. The consumption <strong>of</strong> at least 0.91<br />

kg <strong>of</strong> grain prior to weaning has been recommended (Davis and Drackley, 1998). The failure <strong>of</strong><br />

77


calves to consume sufficient calf starter intake leaves calves vulnerable to hunger (De Passillé et<br />

al., 2010). Rumen development takes several weeks, so if calves are not prepared for weaning,<br />

there is the potential for calves to experience some degree <strong>of</strong> hunger for an extended period <strong>of</strong><br />

time. This would result in calves having poor growth, and suffering from hunger until their<br />

rumen develops sufficiently to allow the calf to absorb nutrients.<br />

In conclusion, there are associations between changes in calf behavior and variability in<br />

post-weaning weight gain. Far lying was associated with decreased weight gain post-weaning.<br />

However, this behavior only occurred in calves from one <strong>of</strong> the nurseries. The motivation for this<br />

behaviour may be a result <strong>of</strong> pre-weaning housing environment but further research to look into<br />

the effect <strong>of</strong> housing on behaviour and growth is required. Calves with the greatest alterations in<br />

behavior in the three days following movement were the most likely to have lower weight gains.<br />

This includes steps per day and a rearing environment dependent alteration in lying time. After<br />

an initial adjustment period, calves with poor post-weaning welfare took fewer steps, and spent<br />

less time lying. In conclusion, the activities <strong>of</strong> calves in the weeks post-weaning is associated<br />

with weight gain during this time period.<br />

78


3.5 REFERENCES<br />

Broom, D. M. and J. D. Leaver. 1978. Effects <strong>of</strong> group-rearing or partial isolation on later social<br />

behaviour <strong>of</strong> calves. Anim. Behav. 26:1255-1263.<br />

Budzynska, M. and D. M. Weary. 2008. Weaning distress in dairy calves: Effects <strong>of</strong> alternative<br />

weaning procedures. Applied Animal Behaviour Science. 112:33-39.<br />

Davis, C. L. and J. Drackley. 1998. The Development, Nutrition, and Management <strong>of</strong> the Young<br />

Calf. 1st ed. Iowa State <strong>University</strong> Press, Iowa.<br />

De Passillé, A. M., B. Sweeney and J. Rushen. 2010. Cross-sucking and gradual weaning <strong>of</strong><br />

dairy calves. Appl. Anim. Behav. Sci. 124:11-15.<br />

Haley, D. B. 2006. The Response <strong>of</strong> 5-Week-Old Nursing Dairy Calves and their Dams to being<br />

Weaned in Two Stages Or Abruptly by Separation. Ph.D., <strong>University</strong> <strong>of</strong> Saskatchewan, .<br />

Haley, D. B., D. W. Bailey and J. M. Stookey. 2005. The effects <strong>of</strong> weaning beef calves in two<br />

stages on their behavior and growth rate. Journal <strong>of</strong> Animal Science. 83, no. 9p. 2205-2214.:.<br />

Jasper, J., M. Budzynska and D. M. Weary. 2008. Weaning distress in dairy calves: Acute<br />

behavioural responses by limit-fed calves. Applied Animal Behaviour Science. 110:136-143.<br />

Jolley, W. R. and K. D. Bardsley. 2006. Ruminant coccidiosis. Veterinary Clinics <strong>of</strong> North<br />

America: Food Animal Practice. 22:613-621.<br />

Raussi, S., A. Boissy, E. Delval, P. Pradel, J. Kaihilahti and I. Veissier. 2005. Does repeated<br />

regrouping alter the social behaviour <strong>of</strong> heifers? Appl. Anim Behav. Sci. 93:1-12.<br />

Schillhorn van Veen, T. W. 1986. Coccidiosis in ruminants. Compendium Food Animal. 8:F52-<br />

F58.<br />

Swanson, J. C. and J. Morrow-Tesch. 2001. Cattle transport: Historical, research, and future<br />

perspectives. J. Anim. Sci. 79:E102-109.<br />

USDA. 2010. Dairy 2007, Heifer Calf Health and Management Practices on U.S. Dairy<br />

Operations, 2007. USDA:APHIS:VS, CEAH., Fort Collins, CO. #550.0110.<br />

Veissier, I., A. Boissy, A. M. dePassille, J. Rushen, C. G. van Reenen, S. Roussel, S. Andanson<br />

and P. Pradel. 2001. Calves' responses to repeated social regrouping and relocation. J. Anim. Sci.<br />

79:2580-2593.<br />

Weary, D. M., J. Jasper and M. J. Hötzel. 2008. Understanding weaning distress. Applied<br />

Animal Behaviour Science. 110:24-41.<br />

79


Table 3.1: Descriptive list <strong>of</strong> observed behaviours<br />

Bunk Behavior (n = 65)<br />

Bunk directed Touching the stanchions <strong>of</strong> the feed bunk<br />

Bunk engaged Head through the bars <strong>of</strong> the stanchion beyond poll but<br />

not lowered<br />

Eating Head through stanchions and lowered<br />

Resting Behavior (n = 44)<br />

Close Lying Lying while touching another calf<br />

Far Lying Lying more than 6 calf lengths from another calf<br />

80


Table 3.2: Associations between activity and Day 7 weight <strong>of</strong> calves in the week post-weaning.<br />

All models included Day 0 weight and nursery Locations in the model as a fixed effect and<br />

group nested within group size as a random effect. All behaviours (steps, lying duration, daily<br />

lying bouts) were assessed independently <strong>of</strong> each other.<br />

Behaviour Pre-Weaning<br />

Period<br />

Number <strong>of</strong><br />

Steps<br />

Total<br />

Lying<br />

Duration<br />

Daily<br />

Lying<br />

Bouts<br />

P = 0.68<br />

P = 0.39<br />

P = 0.21 P = 0.15<br />

Transition Period Plateau Period<br />

0.11 ± 0.03 kg lower /100 step<br />

increase (P < 0.001)<br />

Interaction with Nursery<br />

Location, (P = 0.01; Figure 3.4)<br />

81<br />

0.62 ± 0.11kg higher/ 100<br />

step increase (P < 0.001)<br />

0.81 ± 0.35 kg higher/1 hour<br />

increase (P = 0.03)<br />

0.30 ± 0.13 kg higher/ bout<br />

(P = 0.03)


Number <strong>of</strong> steps per day<br />

7000<br />

6000<br />

5000<br />

4000<br />

3000<br />

2000<br />

1000<br />

0<br />

Pre Pre--<br />

Weaning<br />

-2<br />

-1<br />

Transition<br />

Indicates days calves were were were weighed<br />

weighed<br />

weighed<br />

0<br />

1<br />

2<br />

3<br />

Day relative relative relative to weaning<br />

weaning<br />

weaning<br />

weaning<br />

Figure 3. 3.1: : The average average average average average average number number number number number number <strong>of</strong> steps steps steps steps steps steps steps per per per per per per day day day day day day day day day day day (mean (mean ± ± ± ± ± ± standard standard deviation ) by by by by by by weight gain<br />

gain<br />

gain<br />

gain<br />

gain<br />

gain<br />

gain<br />

gain<br />

gain<br />

gain<br />

category in the first week post<br />

post post-weaning weaning for calves (n (n (n (n (n = 70)<br />

70)<br />

70)<br />

70)<br />

70) following weaning weaning weaning and<br />

and<br />

and<br />

and relocation to a<br />

novel environment<br />

environment<br />

environment.<br />

environment<br />

4<br />

82<br />

5<br />

Plateau<br />

6<br />

7<br />

Weight Gain<br />

Category<br />

8<br />

6.5 kg<br />

kg<br />

9<br />

10<br />

11


Total lying time per day (hours)<br />

20<br />

18<br />

16<br />

14<br />

12<br />

10<br />

8<br />

Pre Pre-<br />

Weaning<br />

-22<br />

-11<br />

0<br />

Transition<br />

1<br />

Indicates days calves calves calves were weighed<br />

2<br />

3<br />

Plateau<br />

Days relative to weaning<br />

Figure 3.2: The total total total total total total lying lying lying lying lying lying duration per per per per per per day day day day day (hours) (mean (mean (mean ± standard standard standard standard standard standard deviation deviation deviation deviation deviation deviation ) ) ) ) ) ) for for for for for for for for for for calves calves calves calves calves calves (n (n (n (n (n (n (n =<br />

=<br />

=<br />

=<br />

=<br />

=<br />

70) following following weaning weaning weaning weaning and<br />

and<br />

and relocation to a novel novel environmen<br />

environmen environment, t, and total lying lying lying duration per per per day day day day day by<br />

by<br />

by<br />

by<br />

by<br />

weight gain<br />

gain category in the first week post post-enrolment enrolment<br />

4<br />

83<br />

5<br />

6<br />

7<br />

8<br />

Weight Gain<br />

Category<br />

9<br />

6.5 kg<br />

10<br />

11


Number <strong>of</strong> lying bouts per day<br />

35<br />

30<br />

25<br />

20<br />

15<br />

10<br />

5<br />

0<br />

Preweaning<br />

Transition<br />

-2 -1 0 1 2 3 4 5 6 7 8 9 10 11<br />

Indicates day calves were weighed<br />

Plateau<br />

Days relative to weaning<br />

Figure 3.3: The average number <strong>of</strong> lying bouts (mean ± standard deviation ) per day for calves<br />

(n = 70) following weaning and relocation to a novel environment, and number <strong>of</strong> lying bouts<br />

per day by weight gain category in the first week post-enrolment<br />

84<br />

6.5 kg


Weight Gain (kgs) for first 7<br />

days post-enrolment<br />

6<br />

5<br />

4<br />

3<br />

2<br />

1<br />

0<br />

12 14<br />

Daily lying time during transition period (hours)<br />

Figure 3.4: LSMEANS (±S.E.) <strong>of</strong> post-weaning weight gain for total lying duration by nursery<br />

location interaction (P = 0.01)<br />

85<br />

Nursery 1<br />

Nursery 2


Weight gain (kgs) first 7 days post-enrolment<br />

4.5<br />

4<br />

3.5<br />

3<br />

2.5<br />

2<br />

1.5<br />

1<br />

0.5<br />

0<br />

-0.5<br />

10 15<br />

Proportion <strong>of</strong> total time at bunk (%)<br />

Figure 3.5: LSMEANS (±S.E.) <strong>of</strong> post-weaning weight gain for time at the bunk (% <strong>of</strong> total<br />

observations) by nursery location interaction (P = 0.03), controlling for Day 0 weight, antibiotic<br />

treatment, and pre-weaning calf starter intake, with random effects for group nested within group<br />

size.<br />

86<br />

Nursery 1<br />

Nursery 2


CHAPTER 4: EVALUATION OF A GRADUATED WEANING PRACTICE ON<br />

BEHAVIOR AND PERFORMANCE OF DAIRY CALVES<br />

4.1 INTRODUCTION<br />

Weaning and relocation to group housing is considered to be a time <strong>of</strong> great stress for<br />

dairy calves. The majority <strong>of</strong> milk-fed calves in the United States are individually housed, and<br />

stress associated with weaning and moving to group housing at weaning can result in increased<br />

susceptibility <strong>of</strong> calves to bovine respiratory disease complex (BRD) (McGuirk, 2008). This<br />

increased risk <strong>of</strong> BRD has been attributed to the many changes occurring at this time including<br />

nutritional, environmental, and social stress, as well as increased exposure to pathogens<br />

(McGuirk, 2008). It has been suggested that separating challenges including vaccination,<br />

dehorning, nutritional changes, as well as moving to new social and physical environments could<br />

reduce susceptibility to clinical disease (Quigley, 1996). One recommendation is to pre-condition<br />

calves by mixing them into small groups after they have been weaned, but before they are moved<br />

into group housing in a novel environment. Comparisons <strong>of</strong> the responses <strong>of</strong> pair-housed calves<br />

with those <strong>of</strong> individually housed calves indicate that pair-housed calves had lower heart rates<br />

during transportation (Lensink et al., 2001) and during the week <strong>of</strong> weaning did not experience a<br />

growth check in contrast to individually housed calves (Chua et al., 2002). However, in these<br />

studies, calves were paired for longer than 6 weeks. For managers that house calves individually<br />

during the pre-weaning period, preconditioning may require additional space in the nursery to<br />

house calves for longer periods <strong>of</strong> time. Therefore, for economic and practical reasons it is<br />

important to determine if housing calves in groups for shorter periods <strong>of</strong> time (e.g. one week)<br />

prior to movement to a new facility will be beneficial for health or growth.<br />

87


The concept <strong>of</strong> separating stressors is based on research investigating the effect <strong>of</strong><br />

multiple concurrent stressors, which has been examined in swine and poultry. Concurrent<br />

stressors were found to produce additive effects in pigs on growth rate (Hyun et al., 1998), feed<br />

efficiency (Hyun et al., 1998), metabolic stress response (Ritter et al., 2009), and activity (Hyun<br />

et al., 1998). In poultry, concurrent stressors were also found to be additive in association with<br />

growth and feed conversion efficiency (McFarlane et al., 1989). These researchers investigated a<br />

variety <strong>of</strong> stressors including thermal, social, disease, and confinement stressors. However, this<br />

concept has not been examined in dairy calves using dairy specific stressors or management<br />

systems.<br />

In a previous study (Chapter 3) higher weight gain in newly weaned and mixed calves<br />

was associated with lower activity, measured by steps per day, and more lying during the first 3<br />

days following weaning, and higher activity and more time spent lying from days 4 to 11. This<br />

suggests that activity patterns may be used as an indicator <strong>of</strong> adaptation to weaning, movement<br />

to a novel environment, and social mixing. Based on these findings, if mixing calves prior to<br />

movement to the environment results in improved adaptation to the environment it was expected<br />

that these calves would take more steps and have a greater lying time compared to calves with no<br />

previous social experience. The objective <strong>of</strong> this experiment was to determine the effect <strong>of</strong><br />

separating post-weaning social mixing and movement into a novel environment on activity<br />

patterns and feed intake. The hypothesis was that calves that are preconditioned to a social<br />

environment by mixing into a group prior to movement to a novel environment will adapt more<br />

easily to the novel environment in comparison to calves that are mixed and moved at the same<br />

time.<br />

88


4.2 MATERIALS AND METHODS<br />

This study was conducted at the Iowa State <strong>University</strong> dairy research farm in Ames,<br />

Iowa. A total <strong>of</strong> 64 calves were enrolled: 53 Holsteins, 7 Jerseys, 3 Brown Swiss and 1 Milking<br />

Shorthorn. All calves were born and raised at the facility under the same management conditions.<br />

4.2.1 Pre-enrolment management<br />

Prior to enrolment, the calves were fed 2L <strong>of</strong> pasteurized whole milk twice daily by<br />

nursery bottle. Ad libitum calf starter (custom calf starter: 23% protein and 4 % fat: Heartland<br />

Co-op, West Des Moines, Iowa) and water were available to all calves. They were housed in a<br />

naturally ventilated nursery barn in individual pens from 1 day <strong>of</strong> age. Weaning was initiated at 4<br />

weeks <strong>of</strong> age by reducing the number <strong>of</strong> daily feedings from 2 to 1. Weaning was complete after<br />

1 week, or calves consumed ≥ 0.5 kg <strong>of</strong> calf starter per day. All calves were monitored daily for<br />

clinical disease and any disease events were recorded. All calves were treated in accordance with<br />

the standard operating procedures <strong>of</strong> the herd, and were examined by a veterinarian prior to<br />

treatment for clinical disease.<br />

Bovine respiratory disease (BRD) was scored by researchers daily using a standardized<br />

scoring system (appendix 4.1; McGuirk, 2005) from 1 to 15, with a score <strong>of</strong> 0 indicating no signs<br />

<strong>of</strong> BRD. A score ≥ 5 was defined as clinical BRD and triggered treatment by veterinarian. Scores<br />

between 1 and 4 were defined as subclinical BRD and calves were monitored but not treated.<br />

4.2.2. Housing<br />

Experimental pens (1.2 x 1.8 meters) were located within the Nursery barn calves were<br />

reared in from birth. All pens had 3 solid sides with spindle gates at the front <strong>of</strong> the pen. All pens<br />

89


were bedded with straw. While housed individually, calves had no physical access to other<br />

calves, but had visual access to calves in pens directly across the alley.<br />

The naturally ventilated barn used as the Novel Environment was located adjacent to the<br />

Nursery barn and contained two experimental pens (3.7 m X 6 m), which were bedded with<br />

straw. Treatment groups were housed separately and had visual, but no physical access to the<br />

other treatment group. No other animals were housed in this barn.<br />

4.2.3. Study Treatment<br />

Calves were randomly assigned to one <strong>of</strong> two treatments, either traditional management<br />

in which they were not mixed prior to movement to a novel environment (Traditional) or mixed<br />

prior to movement to a novel environment (Pre-mixed). Eight calves were enrolled at a time with<br />

4 calves in each treatment group. Enrolment occurred when 8 calves were completely weaned<br />

from milk, at which time all calves were moved to experimental pens (Day -4) within the nursery<br />

barn, which were identical in design to calves’ pre-enrolment housing. Calves were randomly<br />

assigned to treatment and experimental pen using a random number generator. Treatment groups<br />

were balanced for age and breed (Holstein or non-Holstein).<br />

From day- 4 to day 0, all calves were individually housed in experimental pens, within<br />

the nursery barn. On day 0, the partitions between 4 experimental pens were removed creating a<br />

single pen (4.8 m x 1.8 m) which housed the Pre-mixed treatment group from Day 0 to Day 7.<br />

Traditional calves were housed in individual experimental pens until day 7. On day 7, both<br />

treatment groups were moved 40 meters to group pens within a naturally ventilated, novel barn.<br />

All calves within a treatment group were housed in the same pen. Treatment groups were<br />

randomly assigned to 1 <strong>of</strong> 2 pens for each replicate. Each pen measured 14 meters by 22 meters<br />

90


with straw bedding. Calves were housed in the novel barn from day 7 to replicate completion on<br />

day 13.<br />

All calves were fed ad libitum calf starter (custom calf starter: 23% protein and 4 % fat:<br />

Heartland Co-op, West Des Moines, Iowa), water and hay during the study period.<br />

4.2.4. Blood Sampling<br />

Whole blood samples (3mL) were collected from all calves on the replicate on days 0<br />

(before and after social mixing), 1, 7 (before and after movement to a novel environment) and 8<br />

via jugular venopuncture. The order <strong>of</strong> blood collection was randomized at the start <strong>of</strong> each<br />

replicate, balanced for treatment and kept consistent throughout the trial. Samples were collected<br />

on Days 0, 1, 7 and 8 at 2 pm. After serum-samples were collected on Day 0, pen dividers<br />

between the Pre-mixed calves were removed and a second sample was collected from all calves<br />

30 minutes later.<br />

After serum samples were collected on Day 7, both treatment groups were moved to the<br />

new barn and second sample was collected 30 minutes after the last calf entered the new barn.<br />

Blood samples were centrifuged for 20 minutes. Serum was divided between 2 serum tubes and<br />

stored in a -20ºC freezer. Duplicate serum samples were analyzed for cortisol using RIA kits<br />

(Coat-A-Count, Diagnostic Products Corporation, Los Angeles, CA.). The co-efficient <strong>of</strong><br />

variation between the samples was 16.7%, which was determined to be outside the acceptable<br />

range as such these results will not be presented.<br />

91


4.2.5 Data Collection<br />

Calf starter intake was measured daily using an electronic scale. On day -4 calves were<br />

<strong>of</strong>fered 3 kg <strong>of</strong> calf starter per calf. After this initial feeding, fresh calf-starter was <strong>of</strong>fered daily,<br />

and increased by 15% based on their previous day’s consumption. The only exception to this was<br />

the day <strong>of</strong> movement to the novel environment in which calves were <strong>of</strong>fered 20% more than their<br />

intake in the previous 24 hours to account for potential increases in appetite associated with the<br />

move.<br />

Calves were weighed individually on an electronic scale on days -1, 6 and 13. Weight<br />

was recorded to the nearest pound. Data were converted to kilograms using SAS 9.1.<br />

On Day -4, IceTags® (IceRobotics, UK) were attached to the rear metacarpal <strong>of</strong> the last<br />

56 calves on this trial and were removed on day +13. IceTags recorded the number <strong>of</strong> steps<br />

taken, the number <strong>of</strong> lying bouts and total duration <strong>of</strong> time spent lying (minutes) on each day.<br />

Activity data (steps, lying duration and lying bouts) were collapsed into three time periods:<br />

baseline period, pre-mixing period, and novel period. The baseline period was based on the<br />

activity <strong>of</strong> calves on day -2. Day-2 was selected as the baseline period to allow calves 2 days to<br />

adjust to the IceTags on their leg. In addition, no study procedures required interaction with<br />

calves on this day in comparison to day 1 in which calves were weighed. The Pre-mixing period<br />

and Novel Environment Period were based on activity on days 0-5 and 7-11, respectively.<br />

Activity on days -1 and 6 were excluded because <strong>of</strong> the disruptive effect <strong>of</strong> weighing calves.<br />

4.2.6 Statistical Analysis<br />

All statistical analyses were conducted using Statistical Analysis S<strong>of</strong>tware (SAS version<br />

9.1; SAS Institute Inc, Cary, NC, USA). Statistical significance was based on Type III tests for<br />

92


fixed effects with an associated P-value <strong>of</strong> less than 0.10. This cut <strong>of</strong>f was selected since this was<br />

a pilot study. All models were built using backwards stepwise elimination. Normality and the<br />

distribution <strong>of</strong> the residuals were visually assessed for all linear models. The effect <strong>of</strong> BRD at<br />

any time during the trial, categorized as no BRD, subclinical BRD, or clinical BRD, was tested<br />

in all models.<br />

Due to expected breed differences in weight gain, the analysis <strong>of</strong> weight gain was<br />

restricted to purebred Holsteins. The association between study treatment and weight gain during<br />

the Pre-mixing and Novel Environment period was analyzed using linear mixed models (Proc<br />

Mixed) with initial body weight as a fixed effect and replicate as a random effect.<br />

Average calf starter intake was determined at the group level (n = 16) for three time<br />

periods; Baseline (Day -4 to Day -1), Pre-mixing (Day 0 to Day 6), and Novel Environment<br />

period (Day 7 to Day 13). The association between study treatment and average calf starter<br />

intake for the three time periods was assessed using a linear mixed model with a random effect<br />

for replicate. The number <strong>of</strong> calves identified with a BRD score ≥1 within each pen <strong>of</strong> calves<br />

was divided into three categories; no calves with BRD, 1 calf with BRD, and more than 2 calves<br />

within a group with BRD.<br />

Associations between treatment and number <strong>of</strong> steps taken, lying duration and number <strong>of</strong><br />

lying bouts during the different time periods were analyzed using linear mixed models with<br />

repeated measures for individual calf nested within replicate. Normality was assessed using<br />

visual assessment <strong>of</strong> the distribution <strong>of</strong> the residuals (Proc Univariate). To correct for non-<br />

normality, a log transformation was applied to the average number <strong>of</strong> steps per day. Covariates<br />

were selected based on the lowest Akaike information criterion (AIC) for the model.<br />

93


4.3 RESULTS<br />

enrolment.<br />

On average, calves were 49 ± 9 (mean ± standard deviation) days <strong>of</strong> age at the time <strong>of</strong><br />

Treatment had no effect on average daily calf starter intake at the pen level (n = 16)<br />

controlling for time period, and number <strong>of</strong> calves with BRD within a pen as fixed effects and<br />

replicate as a random effect (P = 0.45). Average daily calf starter intake during the Baseline,<br />

Mixing and Novel period was 8.2 ± 0.4 kg, 9.5 ± 0.4 kg, and 10.1 ± 0.4 kg, respectively.<br />

For the Holstein calves on this study (n = 53), the weight gain during the mixing period<br />

was 7.4 ± 0.8 kg and 6.7 ± 0.7 kg for the Premixed and Traditional groups, respectively,<br />

controlling for BRD, nursery pen location, and initial body weight as fixed effect covariates and<br />

replicate as a random effect (P = 0.28).<br />

For the Holstein calves on this study, the weight gain in the post-movement period was<br />

6.4 ± 0.4 kg and 7.1 ± 0.4 for the Premixed and Traditional groups, respectively controlling for<br />

initial body weight and BRD as fixed effects and replicate as a random effect (P = 0.15)<br />

Due to computer errors, activity was only recorded for 18 <strong>of</strong> the 28 Traditional calves and<br />

21 <strong>of</strong> the 28 Premixed calves. The average activity <strong>of</strong> calves during the baseline period was 685<br />

± 428 (mean ± standard deviation) steps, 19 ± 5 lying bouts and 18.2 ± 0.9 hours lying down per<br />

day. During the mixing period, the number <strong>of</strong> steps per day was higher in the Premixed group<br />

relative to the Traditional group (P = 0.03, Figure 4.1). In the novel environment, steps per day<br />

was higher in the Traditional group compared to the Premixed group (P = 0.02). Both Premixed<br />

94


and Traditional calves increased their activity over time (P < 0.001). BRD had no impact on<br />

steps per day (P = 0.82).<br />

The average number <strong>of</strong> lying bouts was 19 ± 5, 19 ± 5 and 20 ± 5 for the Traditional<br />

calves and 20 ± 5, 21 ± 4 and 19 ± 4 for the Premixed calves in the baseline, premix and novel<br />

periods, respectively. The number <strong>of</strong> lying bouts per day did not change by time period (P =<br />

0.37). The average lying number <strong>of</strong> lying bouts per time period was not significantly different by<br />

treatment (P = 0.39). BRD was not significantly associated with the average lying number <strong>of</strong><br />

lying bouts (P = 0.87).<br />

A time period and study treatment interaction was significantly associated with the daily<br />

lying duration per time period (P = 0.004; Figure 4.2). Lying durations for Premixed calves in<br />

the mixing and novel environment time periods were1.6 ± 0.3 and 2.1 ± 0.2 hours less than the<br />

baseline lying time, respectively (P < 0.0001 and P < 0.0001, respectively). There was no<br />

difference in lying duration between the baseline and mixing period for the Traditional calves (P<br />

= 0.85). On average, calves with no BRD, subclinical BRD, and Clinical BRD laid down for<br />

16.9 ± 0.3, 17.4 ± 0.3, and 16.5 ± 0.5 hours per day (P = 0.03).<br />

Subclinical BRD (score between 1 and 4) was observed in 8 Traditional calves and 7<br />

Premixed calves. Clinical BRD (score ≥5) was observed in 3 Traditional calves and 2 Premixed<br />

calves. Clinical BRD scores varied from 5 to 9. Subclinical disease occurred throughout the trial<br />

with the majority occurring in the Premixed time period (Table 4.1). At the group level, 3 pens <strong>of</strong><br />

calves had no cases <strong>of</strong> BRD, 5 pens <strong>of</strong> calves had 1 calf with BRD and 8 pens <strong>of</strong> calves had 2 or<br />

more cases <strong>of</strong> BRD.<br />

95


4.3 DISCUSSION<br />

There was no effect <strong>of</strong> treatment on weight gain, or calf starter intake. The small sample<br />

size and the use <strong>of</strong> multiple breeds within pens may have increased the variability <strong>of</strong> the results<br />

and reduced the ability to detect differences. However, even numerically these results are very<br />

similar, indicating that this treatment has minimal impact on the growth and feed intake <strong>of</strong> calves<br />

with the stressors applied in this study.<br />

In the Mixing period, Pre-mixed calves increased the number <strong>of</strong> steps per day and<br />

decreased their daily lying duration compared to Traditional calves. Since the pre-mixing<br />

resulted in an increase in pen dimension these alterations in behavior is likely a result <strong>of</strong><br />

increased play, exploration <strong>of</strong> a novel environment or increased social interactions. Previous<br />

studies have shown that increased space allowance and social contact is positively associated<br />

with play (Jensen et al., 1998; Jensen and Kyhn, 2000). However, since the increased activity<br />

observed in this study is based solely on pedometry, this increased activity could also be the<br />

result <strong>of</strong> increased displacement and aggression. Anecdotally, both increased play and increased<br />

displacement were observed on the day <strong>of</strong> mixing indicating that both factors may be<br />

contributing to the observed increase in activity. A previous study documented decreased lying<br />

time, increased steps per day, increased aggression and exploratory behavior in the 3 hours<br />

following mixing <strong>of</strong> pair housed 5 week old bull calves (Veissier et al., 2001). The results <strong>of</strong> the<br />

current study are novel since activity for the 6 days following mixing were evaluated.<br />

Following movement to a novel environment, Pre-mixed calves displayed greater lying<br />

duration and fewer steps per day compared to Traditional calves, supporting the hypothesis that<br />

Pre-mixed calves would have a lower behavioral response to movement to a novel environment.<br />

96


The decreased activity <strong>of</strong> the Pre-mixed calves in the novel environment relative to traditional<br />

calves may indicate a less stressful adjustment. However this did not result in greater feed intake<br />

or weight gain. An increase in steps per day and lying time in the 4 to 11 days after weaning and<br />

movement to a novel environment has previously been associated with increased weight gain in<br />

weaned dairy calves (Chapter 3). However, the limitation <strong>of</strong> this comparison is the differences in<br />

the stressors applied to the calves. In the current study calves were moved to a novel<br />

environment at least 4 days after weaning, whereas in the previous study calves were both<br />

weaned and moved to a novel environment at the same time.<br />

There were no differences in the number <strong>of</strong> lying bouts in any <strong>of</strong> the time periods. This<br />

could indicate that this factor is not sensitive to management changes and may require larger<br />

stressors than those in this study to be altered. Alternatively, an environment that is lacking<br />

complexity, such as individual housing, may result in decreased activity due to fewer stimuli.<br />

The movement to a novel environment was <strong>of</strong> very short duration in this study and did not<br />

require transport in a vehicle or haltering. If a greater transportation distance was involved the<br />

observed differences between treatments in the novel environment may have been greater.<br />

In conclusion, the preliminary data from this study suggests that mixing calves in the<br />

nursery prior to moving them to a novel group housed environment does alter their behavior in<br />

the novel environment. Based on these findings further study <strong>of</strong> this management system should<br />

be conducted to determine if these behavioural changes indicate increased or decreased calf<br />

welfare.<br />

97


4.4 REFERENCES<br />

Chua, B., E. Coenen, J. van Delen and D. M. Weary. 2002. Effects <strong>of</strong> pair versus individual<br />

housing on the behavior and performance <strong>of</strong> dairy calves. J. Dairy Sci. 85:360-364.<br />

Hyun, Y., M. Ellis, G. Riskowski and R. W. Johnson. 1998. Growth performance <strong>of</strong> pigs<br />

subjected to multiple concurrent environmental stressors. J. Anim. Sci. 76:721-727.<br />

Jensen, M. B. and R. Kyhn. 2000. Play behaviour in group-housed dairy calves, the effect <strong>of</strong><br />

space allowance. Applied Animal Behaviour Science. 67:35-46.<br />

Jensen, M. B., K. S. Vestergaard and C. C. Krohn. 1998. Play behaviour in dairy calves kept in<br />

pens: The effect <strong>of</strong> social contact and space allowance. Appl. Anim. Behav. Sci. 56:97-108.<br />

Lensink, B. J., S. Raussi, X. Boivin, M. Pyykkönen and I. Veissier. 2001. Reactions <strong>of</strong> calves to<br />

handling depend on housing condition and previous experience with humans. Applied Animal<br />

Behaviour Science. 70:187-199.<br />

McFarlane, J. M., S. E. Curtis, R. D. Shanks and S. G. Carmer. 1989. Multiple concurrent<br />

stressors in chicks. 1. effect on weight gain, feed intake, and behavior. Poult. Sci. 68:865-866.<br />

McGuirk, S. M. 2008. Disease management <strong>of</strong> dairy calves and heifers. Vet. Clin. Food. Anim.<br />

24:139-153.<br />

Quigley, J. 1996. Calf note #16 – stress at weaning. Calf Notes . www.calfnotes.com<br />

Ritter, M. J., M. Ellis, D. B. Anderson, S. E. Curtis, K. K. Keffaber, J. Killefer, F. K. McKeith,<br />

C. M. Murphy and B. A. Peterson. 2009. Effects <strong>of</strong> multiple concurrent stressors on rectal<br />

temperature, blood acid-base status, and longissimus muscle glycolytic potential in marketweight<br />

pigs. J. Anim. Sci. 87:351-362.<br />

Veissier, I., A. Boissy, A. M. dePassille, J. Rushen, C. G. van Reenen, S. Roussel, S. Andanson<br />

and P. Pradel. 2001. Calves' responses to repeated social regrouping and relocation. J. Anim. Sci.<br />

79:2580-2593.<br />

98


Table 4.1: Number <strong>of</strong> calves identified with bovine respiratory disease complex (BRD) by<br />

treatment group and study time period.<br />

Time Period<br />

Treatment BRD category Baseline Period<br />

Traditional<br />

n = 32<br />

Premixed<br />

n = 32<br />

(Day -4 to Day 0) (Day 0 to 6) (Day 7 to 14)<br />

No BRD 31 25 27<br />

99<br />

Mixing<br />

Novel Period<br />

Subclinical<br />

BRD (score 1-4)<br />

1 4 4<br />

Clinical BRD<br />

(score 5-15)<br />

0 3 1<br />

No BRD 32 26 27<br />

Subclinical<br />

BRD (score 1-4)<br />

0 4 5<br />

Clinical<br />

BRD(score 5-<br />

15)<br />

0 2 0


Steps per day<br />

2500<br />

2000<br />

1500<br />

1000<br />

500<br />

0<br />

Figure 4.1: Steps per day (Mean and 95% Confidence Interval ) for each <strong>of</strong> the time periods,<br />

controlling for breed and age as fixed effects, replicate as a random effect and calf nested within<br />

replicate and treatment as a repeated measure.<br />

a,b,c different subscripts within treatments indicates significant differences between time<br />

period within treatment groups (P ≤ 0.10)<br />

a<br />

Premixed<br />

Traditional<br />

P=0.26<br />

a<br />

a<br />

P=0.03<br />

b b c c<br />

Baseline Mixing Novel<br />

Time Period 1<br />

1 Time Periods<br />

Baseline: Activity <strong>of</strong> calves on Day -2, while both Traditional and Premixed calves are<br />

individually housed<br />

Mixing: Activity <strong>of</strong> calves during Days 0-6: Traditional calves are individually housed and<br />

Premixed calves are housed in groups <strong>of</strong> 4.<br />

Novel: Activity <strong>of</strong> calves following movement to a Novel Environment between Day 7-13. Both<br />

Traditional and Premixed calves are housed in groups <strong>of</strong> 4.<br />

100<br />

P=0.02


Average lying time ( hours)<br />

20<br />

18<br />

16<br />

14<br />

12<br />

10<br />

8<br />

6<br />

4<br />

2<br />

0<br />

P=0.05 P=0.02 P=0.07<br />

a a<br />

b<br />

Baseline Mixing Novel<br />

Figure 4.2: LSMeans ( ± S.E.) <strong>of</strong> total daily lying time (hours), controlling for breed and age<br />

with a random effect for replicate and calf nested within replicate and treatment.<br />

a,b: different subscripts indicate a significant difference between time periods with treatments,<br />

(P≤0.10).<br />

a<br />

101<br />

c<br />

b<br />

Premixed<br />

Traditional


Appendix 4.1: Scoring sheet used for health events<br />

0 1 2 3<br />

Rectal Temperature<br />

37.8 to 38.2 o C 38.3 to 38.8 o C 38.9 to 39.4 o C ≤ 39.5 o C<br />

None Induce single<br />

cough<br />

Normal serous<br />

discharge<br />

Small amount <strong>of</strong><br />

unilateral cloudy<br />

discharge<br />

Normal Small amount <strong>of</strong><br />

ocular discharge<br />

Normal Ear flick or head<br />

shake<br />

Normal Semi-formed,<br />

pasty<br />

Cough<br />

Induced repeated<br />

coughs or occasional<br />

spontaneous cough<br />

Nasal Discharge<br />

Bilateral, cloudy or<br />

excessive mucus<br />

discharge<br />

Eyes Scores<br />

Moderate amount <strong>of</strong><br />

bilateral discharge<br />

Ear Scores<br />

Fecal Scores<br />

Slight unilateral<br />

droop<br />

Loose, but stays on<br />

top <strong>of</strong> bedding<br />

102<br />

Repeated<br />

spontaneous coughs<br />

Copious bilateral<br />

mucopurulent<br />

Heavy ocular<br />

discharge<br />

Head tilt or bilateral<br />

droop<br />

Watery, sifts through<br />

bedding<br />

Reference: http://www.vetmed.wisc.edu/dms/fapm/fapmtools/8calf/calf_health_scoring_chart.pdf


CHAPTER 5: THE EFFECT OF TREATMENT WITH LONG-ACTING<br />

ANTIBIOTIC UPON ARRIVAL AT A HEIFER RAISING FACILITY ON DISEASE<br />

AND GROWTH IN COMMERCIAL DAIRY HEIFERS<br />

5.1 ABSTRACT<br />

Commercial heifer rearing facilities face unique disease challenges, and <strong>of</strong>fer an<br />

opportunity to study the impact and prevention <strong>of</strong> these diseases. The objectives <strong>of</strong> this study<br />

were to evaluate a single subcutaneous injection <strong>of</strong> tulathromycin (TUL) in the early postnatal<br />

period, administered upon arrival at a commercial heifer raising facility, on the incidence <strong>of</strong><br />

disease, to describe the risk factors for morbidity and the impact <strong>of</strong> disease on growth, and to<br />

investigate the association <strong>of</strong> Mycoplasma bovis (M. bovis) with the incidence <strong>of</strong> otitis media.<br />

Calves (n = 788) were randomly assigned to study treatment with TUL or a placebo<br />

(CONTROL) upon arrival at the heifer raising facility and were observed for disease daily for 8<br />

weeks by farm staff. Culturing and speciation <strong>of</strong> M. bovis was performed on nasal swabs<br />

collected from a subset <strong>of</strong> calves (n = 66) at 0, 2 and 4 weeks <strong>of</strong> age. Analysis <strong>of</strong> ADG was<br />

conducted using linear mixed models with random effects for source farm, housing location, and<br />

enrolment group nested within housing location. Analysis <strong>of</strong> morbidity was conducted using<br />

logistic regression with source farm, housing location, and enrolment group nested within<br />

housing location as random effects. TUL calves were 60% (OR: 0.41; 95% CI: 0.58-0.82) less<br />

likely to be treated for otitis relative to CONTROL calves (P = 0.002). TUL calves were 43%<br />

(OR: 0.57; 95% CI 0.39-0.84) less likely to be treated for neonatal calf diarrhea complex (P =<br />

0.005). The ADG <strong>of</strong> TUL calves was 0.02 ± 0.01 kg greater than CONTROL calves (P < 0.01).<br />

Failure <strong>of</strong> passive transfer, non-specific fever, bovine respiratory disease complex and neonatal<br />

calf diarrhea complex decreased ADG. Twenty-six percent (n = 85) <strong>of</strong> animals sampled tested<br />

103


positive for M. bovis and 4 different strains were identified. M. bovis was present in this<br />

population but was not associated with otitis media.<br />

104


5.2 INTRODUCTION<br />

There is a growing trend for dairy heifers to be reared <strong>of</strong>f site from milking herds in<br />

commercial heifer raising facilities. In the United States, the proportion <strong>of</strong> operations that sent<br />

calves <strong>of</strong>f-site to commercial heifer raisers increased from 3.6% in 2002 to 4.7% in 2007<br />

(USDA, 2007). In 2007, 11.5% <strong>of</strong> heifers in the United States were from farms that sent their<br />

heifers, primarily pre-weaned calves, <strong>of</strong>f-site to be raised. The commercial rearing industry is an<br />

important focus <strong>of</strong> study since it has unique challenges and opportunities not seen in the<br />

traditional dairy farm model. These farms allow personnel to specialize in the recognition,<br />

management and treatment <strong>of</strong> calf diseases, and have reduced the risk <strong>of</strong> transfer <strong>of</strong> adult cow<br />

diseases to calves after the first week <strong>of</strong> life. This is in contrast to the traditional dairy model, in<br />

which owners or managers must split their time, money and experience between the lactating and<br />

non-lactating animals. However, this commercial rearing model also presents challenges since it<br />

may involve mixing <strong>of</strong> many susceptible animals <strong>of</strong> the same age from multiple source farms.<br />

Mixing <strong>of</strong> animals from multiple sources is a well-documented risk factor for many diseases<br />

(Maunsell and Donovan, 2008). However, the incidence <strong>of</strong> disease in this sector <strong>of</strong> the industry<br />

is not well documented. For instance, participation in the National Animal Health Monitoring<br />

System (NAHMS) survey is based on having at least one milking cow, and as such heifer raising<br />

facilities are not eligible to participate in the study (USDA, 2009).<br />

The administration <strong>of</strong> prophylactic antibiotic has been used to reduce infectious disease<br />

risk at weaning or movement to group housing in dairy replacement heifers (<strong>Stanton</strong> et al., 2010)<br />

and in veal calves, especially during BRD outbreaks(Catry et al., 2008). However, this practice<br />

has been much more thoroughly studied for prevention <strong>of</strong> bovine respiratory disease in beef<br />

105


feedlots (Van Donkersgoed, 1992). In 1999, 83% <strong>of</strong> feedlots used oral prophylactic or<br />

metaphylactic antibiotics and 42% <strong>of</strong> feedlots administered an injectable antibiotic to high-risk<br />

animals to reduce the risk <strong>of</strong> bovine respiratory disease (USDA, 2000). The experience <strong>of</strong> beef<br />

cattle entering the feedlot has many similarities with that <strong>of</strong> dairy calves entering a commercial<br />

heifer raising facility. Both animal types are vulnerable to infection as they are exposed to<br />

disease-causing organisms from multiple sites. Beef cattle are likely to have been recently<br />

weaned, and may have been subjected to a lengthy transportation period which could include<br />

transit through 1 or more sales barns. These stressors are <strong>of</strong>ten associated with<br />

immunosuppression (Carroll and Forsberg, 2007). The neonatal dairy calf has a naïve immune<br />

system and relies primarily on colostral antibodies received from the dam. This is a high risk<br />

period for calves; almost one quarter <strong>of</strong> all calves develop diarrhea in the pre-weaning period,<br />

and 12% <strong>of</strong> calves experience a respiratory disease event (USDA, 2010). Combining this<br />

susceptibility with mixing <strong>of</strong> animals from multiple sources may put animals at increased risk <strong>of</strong><br />

morbidity and mortality. To date, no study has investigated the effectiveness <strong>of</strong> a single injection<br />

<strong>of</strong> a long-acting antibiotic, administered upon arrival at a commercial heifer raising facility, for<br />

the control and prevention <strong>of</strong> clinical disease.<br />

Tulathromycin (Draxxin, Pfizer Animal Health Group, New York, NY) is a tiamilide<br />

compound, a subclass <strong>of</strong> macrolide long-acting antimicrobials. It is distinguished by maintaining<br />

therapeutic concentrations against pathogens commonly associated with Bovine Respiratory<br />

Disease complex for up to 10 d (Nowakowski et al., 2004). Tulathromycin is approved for<br />

treatment and control <strong>of</strong> BRD in beef cattle and nonlactating dairy cattle. Tulathromycin has<br />

106


een shown to be effective in the treatment <strong>of</strong> BRD in 3- to 9-wk-old calves challenged with<br />

Mycoplasma bovis (Godinho et al., 2005; Godinho et al., 2007).<br />

The primary objective <strong>of</strong> the study was to evaluate a single subcutaneous injection <strong>of</strong><br />

tulathromycin (TUL) in the early postnatal period administered upon arrival at a commercial<br />

heifer raising facility, on the incidence <strong>of</strong> disease in young dairy heifers. A second objective was<br />

to describe the risk factors for morbidity and the impact <strong>of</strong> disease on growth <strong>of</strong> calves, in a<br />

commercial heifer raising facility. The third objective was to investigate the association <strong>of</strong><br />

Mycoplasma bovis (M. bovis) with otitis media. The primary hypothesis was that a single<br />

injection <strong>of</strong> TUL upon arrival would reduce the prevalence <strong>of</strong> M. bovis and bacterial pathogens<br />

and reduce the spread <strong>of</strong> these pathogens to other animals and so lower the incidence <strong>of</strong> clinical<br />

otitis, diarrhea and non-specific fever in the first 8 weeks <strong>of</strong> life. The secondary hypothesis was<br />

that failure <strong>of</strong> passive transfer (FPT) and the occurrence <strong>of</strong> diarrhea, non-specific fever, bovine<br />

respiratory disease (BRD), and otitis media would negatively affect average daily gain (ADG).<br />

5.3 MATERIALS AND METHODS<br />

5.3.1 Heifer Management and Housing<br />

All calves were housed at a commercial heifer raising facility in western New York State.<br />

Enrollment <strong>of</strong> 788 heifer calves occurred between October 2007 and June 2008. Calves came<br />

from 5 source farms in western New York State. The source farms contributed between 43 and<br />

315 calves to the trial. Calves were between 0 and 10 days <strong>of</strong> age at arrival at the farm, with a<br />

median age <strong>of</strong> 3 days. Upon arrival at the facility, each calf was processed according to a<br />

standard farm protocol, and was then moved to individual housing. Individual housing was<br />

available in 6 naturally ventilated nursery barns and 47 individual outdoor hutches. Farm<br />

107


protocol included the identification <strong>of</strong> any congenital abnormalities and measurement <strong>of</strong> serum<br />

total protein with a digital refractometer. All calves tested negative for persistent infection with<br />

Bovine Viral Diarrhea (BVD) upon arrival at the farm. Calves were vaccinated at 2 and 4 weeks<br />

<strong>of</strong> age with an intramuscular modified live vaccine against Infectious Bovine Rhinotracheitis,<br />

BVD Type 1 and Type 2, Bovine Respiratory Syncytial Virus, and Parainfluenza-3 Virus (Vista-<br />

5, Intervet Schering-Plough Animal Health, Boxmeer, The Netherlands).<br />

Calves were fed 4L <strong>of</strong> milk replacer (Excel Calf Milk Replacer 26/18, Grober Nutrition,<br />

Cambridge, Ontario, Canada) twice daily (total <strong>of</strong> 8L/day) and had access to ad libitum calf<br />

starter (23% percent protein;5% fat) and water. Weaning from milk replacer was initiated at 5<br />

wks by decreasing to one milk replacer feeding per day and was complete at 6 weeks <strong>of</strong> age.<br />

Calves were moved from their individual pens or hutches to group housing 8 weeks after<br />

arrival/enrollment, which represented the end <strong>of</strong> this trial.<br />

5.3.2 Study Treatments<br />

Calves were randomly assigned to treatment with tulathromycin (TUL) or a placebo as a<br />

negative control (CONTROL) using a random number table and balancing allocation in blocks<br />

<strong>of</strong> 20 based on the order they were processed by farm staff upon arrival at the heifer raising<br />

facility. Researchers and barn staff were blind to treatment. All TUL calves were injected with<br />

1.0 mL containing 100 mg <strong>of</strong> tulathromycin (Draxxin®, Pfizer, Animal Health Group, New<br />

York, New York, USA) by subcutaneous injection on the day <strong>of</strong> arrival at the heifer raising<br />

facility. The placebo was a 1 mL visually identical mix <strong>of</strong> propylene glycol and sterile saline.<br />

108


5.3.3 Outcome Measures<br />

All calves were monitored twice daily for signs <strong>of</strong> clinical disease by trained barn staff<br />

using standardized disease definitions (Table 5.1). The incidences <strong>of</strong> disease in this study were<br />

calculated based on the number <strong>of</strong> new cases <strong>of</strong> a defined disease over the study period (0-8<br />

weeks) (Dohoo et al., 2003). Failure <strong>of</strong> passive transfer (FPT) was determined based on serum<br />

total protein < 5.4 g/dL (Tyler et al., 1998). The established farm practice was to record all<br />

treatments on standardized forms at the time <strong>of</strong> treatment and these data were later entered into a<br />

farm management computer program (DairyComp 305, Valley Agriculture S<strong>of</strong>tware Inc., Tulare,<br />

CA). In addition to the routine diseases recorded, farm staff were additionally asked to indicate if<br />

calves affected with ear droop showed signs <strong>of</strong> unilateral ear droop (UNI), bilateral ear droop<br />

(BIL), or if the head was held in an unusually tilted position (Figure 5.1).<br />

All calves were weighed on an electronic scale (Tru-test - ID3000, Tru-Test Incorporated,<br />

Mineral Wells Texas) and measured for height at the withers (to the nearest inch) using a<br />

yardstick upon arrival at the farm (study enrollment) and at exit from the nursery facilities 8<br />

weeks later (study completion).<br />

5.3.4 Sampling for Mycoplasma testing<br />

Nasal swabs were collected from a subset <strong>of</strong> study calves. Every other week, the<br />

youngest five calves with no history <strong>of</strong> ear droop were enrolled by study personnel for<br />

mycoplasma sampling and were re-sampled two and four weeks later. Additionally, nasal<br />

samples were collected twice a month from clinically affected and healthy calves. At each visit,<br />

two calves with that were identified and treated for ear droop by farm staff within 24 hours<br />

109


(cases) and two calves closest in age to the affected calves, but with no history <strong>of</strong> ear droop, were<br />

sampled. Case and control calves were not re-sampled.<br />

Study personnel wearing sterile disposable gloves used swabs moistened with<br />

mycoplasma enrichment medium to swab the nares <strong>of</strong> selected calves. Swabs were placed in<br />

enrichment medium and shipped to the Mastitis Research Laboratory at Washington State<br />

<strong>University</strong> in Pullman, Washington. Specimens were incubated at 37 o C in 10% CO2 for 4 days.<br />

On the fourth day, 100 µL <strong>of</strong> enrichment medium was streaked on a Mycoplasma agar plate<br />

(Hogan et al., 1999). Plates were incubated at 37 o C in 10% CO2 for 7 to14 days, then examined<br />

with a 15X dissecting microscope for colonies with the distinctive “fried-egg” appearance<br />

(Hogan et al., 1999). Results were considered positive if any Mycoplasma colonies were seen<br />

and negative in the absence <strong>of</strong> Mycoplasma growth. Speciation was done using PCR analysis,<br />

and fingerprinting for type was performed using pulsed- field electrophoresis patterns <strong>of</strong> the<br />

positive isolates as described by Biddle et al, ((Biddle et al., 2005).<br />

5.3.5 Statistical Analysis<br />

All analyses were performed using SAS® statistical s<strong>of</strong>tware (SAS Institute Inc., Cary,<br />

NC) v 9.1, with statistical significance based on Type III test for fixed effects having a P-value ≤<br />

0.05. All models were built using backward step-wise elimination. Normality was assessed for<br />

all linear models.<br />

The associations between disease events and ADG and height at study completion were<br />

analyzed using a linear mixed model (PROC MIXED) with random effects for source farm,<br />

location <strong>of</strong> barn or hutch system, and cohort group nested within housing location. A cohort<br />

group was defined as animals in the same housing system at the same time. Disease outcomes<br />

110


were neonatal calf diarrhea complex, otitis (bilateral and unilateral), non-specific fever, non-<br />

specific lameness, navel ill, and bovine respiratory disease (BRD). The associations <strong>of</strong> study<br />

treatment, difficulty with the calf’s birth, housing type (hutch versus barn) and FPT with growth<br />

were tested in separate models from those that examined the association <strong>of</strong> growth with disease<br />

events because disease events are likely to be intervening variables in the path from risk factors<br />

to decreased growth and as such cannot be included in the same model. Models <strong>of</strong> ADG and<br />

height included weight and height at arrival, respectively and age at <strong>final</strong> measurement as fixed<br />

effects.<br />

Potential predictors <strong>of</strong> first treatments for neonatal calf diarrhea complex, unilateral ear<br />

droop, bilateral ear droop and fever were evaluated using generalized linear mixed models with a<br />

logit transformation (PROC GLIMMIX). Source farm, location <strong>of</strong> barn or hutch system, and<br />

cohort group nested within housing system location were included as random effects. A cohort<br />

group was defined as all animals in the same housing system at the same time. Models <strong>of</strong><br />

unilateral and bilateral ear droop were built separately since it is not known if the unilateral and<br />

bilateral ear droop have the same risk factors. The model for unilateral ear droop included calves<br />

with unilateral ear droop versus calves with no recorded history <strong>of</strong> any ear droop, while the<br />

model for bilateral ear droop included calves with bilateral ear droop versus calves with no<br />

recorded history <strong>of</strong> any ear droop. Study treatment, birth difficulty score, housing system type<br />

(hutch versus barn) and FPT were tested as predictors in all disease models.<br />

The relationship between study treatment and mortality was evaluated using a chi-square<br />

test (PROC FREQ).<br />

111


No statistical analysis was performed to evaluate the association between source farm and<br />

Mycoplasma due to the low number <strong>of</strong> positive samples.<br />

5.4 RESULTS<br />

At enrolment, calves were 3 ± 2 days old (mean ± standard deviation), weighed 41.2 ±<br />

4.4 kg and were 80±3 cm tall at the withers.<br />

5.4.1 Health Events in the Study Population<br />

The most frequent health conditions were neonatal calf diarrhea complex (84%) and otitis<br />

media (78%) (Table 5.2).<br />

Unilateral and bilateral ear droop were observed in 4% (32/788) and 52 % (414/788) <strong>of</strong><br />

calves, respectively. Farm staff failed to record whether ear droop was unilateral or bilateral at<br />

first treatment for 170 <strong>of</strong> the 616 calves treated for ear droop. Analyses <strong>of</strong> ear droop excluded<br />

calves that did not have a record <strong>of</strong> the numbers <strong>of</strong> ears affected. None <strong>of</strong> the calves showed<br />

signs <strong>of</strong> ataxia during this trial. Serum total protein was not measured for 10 calves on this trial<br />

but the prevalence <strong>of</strong> FPT (total protein ≤ 5.4) was only 5% overall.<br />

5.4.2 Morbidity<br />

Age, weight and height at enrolment were not significantly different between TUL and<br />

CONTROL calves. The incidence <strong>of</strong> diarrhea was 87% (345/395) and 80% (314/393) in the<br />

CONTROL and TUL calves, respectively. CONTROL calves were 1.8 (95% Confidence<br />

Interval: 1.2-2.6) times more likely to have neonatal calf diarrhea complex than calves that<br />

received TUL, (P = 0.005), controlling for source farm, housing location and group nested within<br />

112


housing location as random effects. There was a tendency for calves from the nursery barn to be<br />

at greater risk for diarrhea compared to calves from the hutches {OR: 2.4: 0.9-6.4; P = 0.08}.<br />

The incidence <strong>of</strong> unilateral, bilateral and unrecorded ear droop were 3% (11/393), 49%<br />

(193/393) and 22% (86/393) in TUL calves, and 5% (21/395), 56% (221/395) and 21% (84/395)<br />

in CONTROL calves, respectively. The probability <strong>of</strong> TUL calves being treated for otitis (either<br />

unilateral or bilateral ear droop) was 60% lower (OR: 0.41: CI: 0.60-0.82) than CONTROL<br />

calves (P = 0.002) controlling for source farm, housing location and group nested within housing<br />

location as random effects. Otits was observed in 5% (1/19) <strong>of</strong> calves in the hutches and 17%<br />

(31/185) <strong>of</strong> calves in the nursery barns (P = 0.006). Study treatment was the only variable that<br />

was significant in the <strong>final</strong> models for unilateral and bilateral ear droop that controlled for group<br />

and source farm as random effects. TUL calves were 0.26 (CI: 0.11-0.65) times as likely to<br />

display unilateral ear droop relative to CONTROL calves (P = 0.03). The probability <strong>of</strong> TUL<br />

calves displaying bilateral ear droop was 40% lower (OR: 0.58: CI: 0.40-0.83) than CONTROL<br />

calves (P = 0.003).<br />

The incidence <strong>of</strong> Non-Specific Fever (NSF) in the TUL group was 13% (52/393) and<br />

18% (71/395) in the CONTROL group. Calves in the CONTROL group tended to have greater<br />

odds <strong>of</strong> having NSF compared to calves in the TUL group (OR: 1.5; CI: 1-2.2; P = 0.07),<br />

controlling for source farm, housing location and group nested within housing location as<br />

random effects. There was no difference in the incidence <strong>of</strong> NSF between hutch and nursery barn<br />

housed calves (P = 0.12). The incidence <strong>of</strong> BRD was 2% (9/393) and 3% (12/395) in TUL and<br />

CONTROL groups, respectively (P = 0.52). The incidence <strong>of</strong> lameness was low with 0.8%<br />

(6/393) and 1.3% (10/395) in the TUL and CONTROL group, respectively (P = 0.32).<br />

113


5.4.3 Growth<br />

Three calves returned to their source farm at the end <strong>of</strong> the study without being weighed<br />

and were excluded from analyses <strong>of</strong> ADG and height. Prior to the end <strong>of</strong> the trial, 14 calves died<br />

and were excluded from analyses <strong>of</strong> ADG and height.<br />

At the end <strong>of</strong> the study, calves were 58 ± 7 (mean ± standard deviation) days old and<br />

weighed 75 ± 11 kg. Average daily gain over the study period was 0.61 ± 0.13 kg/d.<br />

Relationship between morbidity and average daily gain<br />

Diarrhea, lameness, BRD and NSF were significantly associated with ADG, controlling<br />

for initial body weight and age at <strong>final</strong> measurement as fixed effects and source farm, group and<br />

group within barn as random effects. Diarrhea was associated with a decrease in ADG <strong>of</strong> 0.03 ±<br />

0.01 kg relative to non-diarrheic calves (P = 0.01). NSF was associated with a decrease in ADG<br />

<strong>of</strong> 0.07 ± 0.01 kg relative to calves that were not diagnosed with NSF (P < 0.001). Calves with<br />

lameness had a decrease in ADG <strong>of</strong> 0.16 ± 0.03 relative to non-lame calves (P


model (P = 0.54). The ADG <strong>of</strong> calves in hutch housing were not different than calves housed in<br />

barns (P = 0.79).<br />

5.4.4 Height at movement to group housing<br />

On average, calves were 92 ± 3 (mean ± standard deviation) cm tall at the end <strong>of</strong> the<br />

study, an average growth <strong>of</strong> 11 ± 3 cm over the course <strong>of</strong> the trial.<br />

Relationship between disease and height<br />

Diarrhea, otitis and NSF were associated with decreased height upon exit from the<br />

nursery, controlling for arrival height as a fixed effect, and source farm, housing location and<br />

group nested within housing location as random effects. Calves with diarrhea were shorter by 1 ±<br />

0.2 cm relative to calves without diagnosed diarrhea, (P = 0.01). Calves with unilateral ear droop<br />

were shorter by 2 ± 0.5 cm relative to calves without diagnosed ear droop, (P = 0.001). Calves<br />

with bilateral and unrecorded ear droop did not differ in height from unaffected calves, (P = 0.32<br />

and 0.12, respectively). Calves with NSF were 1 ± 0.2 cm shorter relative to calves without<br />

diagnosed NSF, (P < 0.001).<br />

Relationship between disease predictors and height<br />

Failure <strong>of</strong> passive transfer was a significant predictor <strong>of</strong> height at movement from the<br />

nursery barn. Controlling for height at arrival and age at measurement, with random effects to<br />

control for source farm, housing location and group nested within housing location, calves with<br />

FPT were 1.2 ± 0.4 cm smaller than calves with successful passive transfer (P < 0.001). There<br />

was a tendency for TUL calves to be 0.3 ± 0.2 cm taller at exit from the study (P = 0.08). There<br />

was no difference in height between calves housed in hutches and calves housed in barns (P =<br />

0.20).<br />

115


5.4.4 Testing for Mycoplasma<br />

Of the 66 calves that were enrolled upon arrival at the heifer raising facility and sampled<br />

at 0, 2, and 4 weeks, 17 (26%) calves had one positive test for Mycoplasma. Of the sampled<br />

calves, 89% (59/66) were treated for neonatal calf diarrhea and 74% (49/66) had ear droops<br />

during the study (Table 5.3). No samples collected in week 0 were positive for Mycoplasma.<br />

Positive samples were collected from two calves in week 2 and from 15 calves in week 4. Of the<br />

positive calves, 9 out <strong>of</strong> 17 were from the TUL group. Ear droop was recorded in 10 out <strong>of</strong> 17 <strong>of</strong><br />

the calves that had positive samples collected. Of the 10 calves with ear droop, 90% were<br />

diagnosed with ear droop between week 2 and week 4 samples. Positive cultures were found in<br />

calves from 4 out <strong>of</strong> the 5 source farms (Table 5.4). The fifth farm was a small facility and only 2<br />

calves were cultured from this farm during the study.<br />

M.bovis samples were strain typed and 5 distinct strains identified with 4 out <strong>of</strong> 5 source<br />

farms having at least 2 strains. Due to the fragility <strong>of</strong> M. bovis, there was not enough DNA to<br />

complete fingerprinting on 5 samples, and as such, these samples could only be confirmed as M.<br />

bovis. In addition, 4 samples did not survive frozen storage prior to confirming strain <strong>of</strong><br />

fingerprinting and can only be confirmed as Mycoplasma species.<br />

Nasal samples were taken from 42 cases and 43 control animals. One <strong>of</strong> the active cases<br />

from source farm 1 and 1 <strong>of</strong> the control animals from source farm 4 tested positive for M. bovis.<br />

The case from farm 1 was identified as strain A. The control animal showed signs <strong>of</strong> ear droop<br />

within 24 hours <strong>of</strong> the positive sample. The strain M. bovis from farm 4’s case did not have<br />

enough DNA to complete fingerprinting, and as such, this sample could only be confirmed as M.<br />

bovis.<br />

116


5.4.4 Mortality<br />

Eight calves from CONTROL and 6 calves from TUL died during the trial for a mortality<br />

risk to 8 weeks <strong>of</strong> age <strong>of</strong> 2% (14/788). The main causes <strong>of</strong> mortality as determined by farm staff<br />

were diarrhea (n = 3), bloat (n = 2), high fever (n = 2) and chronic poor health (n = 2). Other<br />

causes included paralysis, stomach ulcer, and swelling <strong>of</strong> neck. No deaths were directly<br />

associated with ear droop.<br />

5.5. DISCUSSION<br />

Disease pressures and management decisions that affect calf disease are likely to be<br />

different on a commercial heifer raising facility than on a traditional farm which raises its own<br />

replacement animals. Heifer raising facilities are likely to be more focused on monitoring <strong>of</strong><br />

growth and disease than traditional farms based on their need to report and justify the cost <strong>of</strong><br />

rearing calves to the contracted source farms; in addition their risk factors are different as a result<br />

<strong>of</strong> completely isolating calves from older animals but exposing them to animals from multiple<br />

sources. In comparison to published reports, this study herd had a much higher incidence <strong>of</strong><br />

diarrhea, a similar incidence <strong>of</strong> respiratory disease and navel ill and a much lower incidence <strong>of</strong><br />

hernia and FPT, (Table 5.5). These differences are likely due to a combination <strong>of</strong> factors that<br />

make commercial heifer raising facilities unique. For diarrhea, the high level <strong>of</strong> vigilance by the<br />

farm staff and the unrestricted definition <strong>of</strong> diarrhea to any observation <strong>of</strong> s<strong>of</strong>t feces lacking in<br />

form likely increased the apparent incidence. The transport and mixing <strong>of</strong> multiple animals<br />

shortly after birth would also increase disease pressure. This study farm defined non-specific<br />

fever as dull and listless with a rectal temperature above 39.5C. This definition lacks symptoms<br />

specific to BRD, and as such may either represent a response to another unrelated infection or<br />

117


may be an early sign <strong>of</strong> respiratory disease. As such, the decision was made not to lump BRD<br />

and fever together in analysis. However, when the incidences <strong>of</strong> BRD and fever are combined<br />

they are similar to the incidences reported in the literature. In this herd, navel ill and hernias were<br />

on the low end <strong>of</strong> reported incidences. Failure <strong>of</strong> passive transfer was very low for these calves<br />

in comparison to the literature. This is likely due to the incentive and education program <strong>of</strong>fered<br />

by the manager <strong>of</strong> the facility which educates producers on colostrum management and penalizes<br />

FPT by increasing the cost <strong>of</strong> rearing calves with FPT. The risk <strong>of</strong> mortality <strong>of</strong> pre-weaned<br />

heifers in the present study is substantially lower in comparison with the NAHMS Survey, which<br />

recently documented mortality <strong>of</strong> 7.8% in pre-weaned calves (USDA, 2010).<br />

In the 2007 survey, NAHMS reported that the median weight <strong>of</strong> Holstein heifer calves<br />

between 56 and 62 days <strong>of</strong> age in the United States was 80 kg (USDA, 2010). The calves in this<br />

study were slightly below the median calf weight, with an average weight at 8 weeks <strong>of</strong> 75.4 kg.<br />

BRD, fever, diarrhea and lameness affected ADG in this study. BRD and fever were found to<br />

reduce ADG by 0.10 and 0.07 kg, respectively. The reduced impact <strong>of</strong> fever relative to BRD and<br />

the similarity between the two measurements does support our theory that fever is an earlier<br />

stage <strong>of</strong> BRD. The impact <strong>of</strong> these health conditions on ADG has been documented in two other<br />

studies. Virtala et al (1996) found that ADG decreased by 0.07 kg for calves with BRD in the<br />

first month <strong>of</strong> life and Donovan et al (1998a) found that BRD decreased ADG by 0.01 kg in the<br />

first 6 months. The 6-month follow up period <strong>of</strong> the Donovan et al (1998a) study allowed for<br />

factors such as recovery from BRD, other diseases, diet, social interactions, and movement to<br />

group housing to affect the growth <strong>of</strong> calves. Therefore the effect <strong>of</strong> BRD was likely diluted.<br />

118


The present study found that diarrhea decreased ADG by 0.03kg (range <strong>of</strong> 0.01 to 0.05<br />

kg). This is consistent with the findings <strong>of</strong> Donovan et al (1998a) who found that diarrhea<br />

decreased ADG by 0.013 kg. However, these two studies were in conflict with the finding <strong>of</strong><br />

Virtala et al (1996) <strong>of</strong> no effect <strong>of</strong> diarrhea. This may be due to the methods used to measure<br />

weight gain in the latter study. The studies which found an effect <strong>of</strong> ADG measured weight gain<br />

using electronic scale while the one which did not estimated weight gain using weight tape<br />

measurement. The measurement error introduced from using this measurement may have masked<br />

the effect <strong>of</strong> diarrhea.<br />

Otitis did not have an effect on ADG. This is inconsistent with Sockett, et al (Sockett et<br />

al., 2008) who found a 4.68kg decrease in body weight gain in calves with clinical otitis media.<br />

However, the current findings are consistent with Stipkovits et al (Stipkovits et al., 2005), in<br />

which an experimentally induced Mycoplasma bovis infection did not have a significant effect<br />

on ADG during the 18 days following infection when treated with antibiotics. These differences<br />

in the reports on the effect <strong>of</strong> otitis media on ADG may be due to several factors. No statistical<br />

methods were reported in Sockett et al (2008), so comparisons between studies are limited. Early<br />

detection <strong>of</strong> disease is associated with a decrease in the severity <strong>of</strong> signs <strong>of</strong> other diseases<br />

(McGuirk, 2008), and the high vigilance <strong>of</strong> farm staff and early treatment in the present study<br />

may have minimized the effect <strong>of</strong> otitis on average daily gain. Alternatively, our definition <strong>of</strong><br />

otitis needed to be very simple since this was a field study on a commercial farm and did not<br />

allow for advanced diagnostics. Given that our definition <strong>of</strong> otitis was based on ear droop, it is<br />

possible that some <strong>of</strong> the observed bilateral ear droops may have been associated signs <strong>of</strong><br />

depression from other diseases such as diarrhea and may have been an indicator <strong>of</strong> malaise. The<br />

119


small number <strong>of</strong> unilateral ear droops limited our ability to determine the impact <strong>of</strong> this<br />

condition on average daily gain. However, given that there was a tendency for unilateral ear<br />

droop to reduce ADG, further research into the effect <strong>of</strong> this condition on ADG is warranted.<br />

Failure <strong>of</strong> passive transfer decreased ADG by 0.04 kg. This is consistent with Virtala et<br />

al (1996), who found that FPT decreased ADG by 0.05 kg during the first month <strong>of</strong> life. TUL<br />

increased ADG in this study by 0.02 kg. Both passive transfer <strong>of</strong> antibodies and TUL decreased<br />

the risk <strong>of</strong> disease. Calves with FPT are more susceptible to diseases that negatively affect ADG<br />

(Donovan et al., 1998). Tulathromycin is an antibiotic that is primarily targeted against<br />

respiratory disease. However, this study demonstrated a beneficial impact <strong>of</strong> TUL treatment on<br />

the incidence <strong>of</strong> neonatal calf diarrhea complex, perhaps by reducing the incidence <strong>of</strong> BRD and<br />

otitis media, thereby enabling calves to mount a stronger immune response against diarrhea-<br />

causing organisms without the additional burden <strong>of</strong> other diseases. There was no interaction<br />

between TUL and successful passive transfer, with both conferring protection against disease.<br />

This may be due to the low number <strong>of</strong> calves with FPT in this population limiting the power to<br />

investigate interactions. In this study herd, TUL reduced the incidence <strong>of</strong> both unilateral and<br />

bilateral ear droop, diarrhea and tended to reduce the incidence <strong>of</strong> non-specific fever. In a similar<br />

study, Sockett et al (2008) also found that tulathryomycin, administered prophalactically,<br />

reduced the incidence <strong>of</strong> otitis media in milk-fed calves. However, tulathromycin was<br />

administered at 1 and 7 days post arrival, whereas we used a single treatment. If an odds ratio is<br />

calculated from the results reported by Sockett et al (2008), the incidence <strong>of</strong> otitis media was 3.8<br />

(CI: 1.8-9.8) times greater in placebo calves compared to calves that received tulathromycin on<br />

day 1 and 7. This finding is similar to our finding that CONTROL calves were 3.7 (CI: 1.6 - 9.1)<br />

120


and 1.7 (CI: 1.2-2.5) times more likely to display unilateral and bilateral ear droop compared to<br />

calves that received TUL, respectively. The results from these two studies suggest that a single<br />

injection <strong>of</strong> tulathromycin reduces the incidence <strong>of</strong> otits in neonatal calves; further study is<br />

warranted to confirm these findings.<br />

The results from the cultures indicated that M. bovis was present in this herd and<br />

that calves were more likely to test positive for M. bovis when they were greater than 2 weeks <strong>of</strong><br />

age. This may indicate that calves were not contracting Mycoplasma until after they arrived at<br />

the heifer raising facility, or they were not shedding this organism until they were older.<br />

However, clinical otitis media was primarily identified in calves less than 2 weeks <strong>of</strong> age.<br />

There are two potential explanations for the lack <strong>of</strong> association between M. bovis and<br />

otitis. The first is that M. bovis is not an important pathogen in clinical otitis media in young<br />

dairy calves and the second is that due to the sampling schedule and antibiotic use on this farm<br />

false negative were generated. Prior research has found that M. bovis is associated with<br />

respiratory disease, arthritis and otitis (as reviewed by Maunsell and Donovan, 2009).<br />

However, while M. bovis can result in clinical disease in a challenge study, in the field<br />

this relationship is much more complex. Since the farm treatment protocol that required the staff<br />

to administer antibiotics at the time <strong>of</strong> diagnosis, which likely inhibited culture growth since<br />

samples were collected 2-12 hours after treatment, a higher rate <strong>of</strong> false negatives is probable.<br />

Based on the findings <strong>of</strong> this study, we cannot determine which <strong>of</strong> these explanations is accurate.<br />

The use <strong>of</strong> mass antibiotic treatments can be controversial and spark worries about<br />

antimicrobial resistance. This is a valid concern and as such the use <strong>of</strong> antibiotics must be fully<br />

121


evaluated for efficacy and carefully targeted for rational use. Research into the efficacy <strong>of</strong><br />

strategic antibiotic treatments is important to prevent producers from using antibiotics<br />

ineffectively. In order to minimize the risk <strong>of</strong> antimicrobial resistance the antibiotics should be<br />

used strategically on high risk animals within a confined time period <strong>of</strong> high risk. This method <strong>of</strong><br />

antibiotic treatment should not be a blanket treatment undertaken for the purpose <strong>of</strong> growth<br />

promotion but rather a strategic intervention to prevent the outbreak and spread <strong>of</strong> disease in a<br />

highly vulnerable population. In conjunction with this method efforts should be made to reduce<br />

or minimize the risk factors that are manageable such as cleanliness, ventilation, and colostrum<br />

management which can lead to increased disease susceptibility with the aim <strong>of</strong> reducing or<br />

eliminating the need for this program. For this to occur, good record keeping is needed in order<br />

to identify the risk factors and appropriate conditions to use this method and to monitor the<br />

success <strong>of</strong> this program. The ideal circumstance for metaphylactic use <strong>of</strong> anti-infectives is one in<br />

which the risk <strong>of</strong> developing the targeted disease is high in a constrained period <strong>of</strong> time and<br />

treatment success <strong>of</strong> clinical disease is limited, representing an animal welfare and production<br />

concern. Mycoplasma bovis contracted at a source farm would fit these criteria based on the high<br />

incidence <strong>of</strong> disease following movement to the heifer raising facility and the difficulty <strong>of</strong><br />

treating Mycoplasma bovis associated disease (as reviewed by Caswell and Archambault, 2007).<br />

However, based on the previously listed criteria, the lack <strong>of</strong> association between study treatment,<br />

or clinical otitis and M. bovis does suggest that on this farm otitis media does not fit the criteria<br />

for the use <strong>of</strong> metaphylaxis.<br />

In conclusion, TUL decreased the incidence <strong>of</strong> diarrhea, unilateral and bilateral ear droop<br />

in the early post-natal period. In addition, TUL tended to decrease the incidence <strong>of</strong> non-specific<br />

122


fever. Non specific fever, respiratory disease, FPT, lameness and neonatal calf diarrhea complex<br />

had negative effects on average daily gain. Finally, despite 26% <strong>of</strong> sampled animals testing<br />

positive for M. bovis confirming the presence <strong>of</strong> this organism at the facility, this organism was<br />

not associated with otitis. With improved culturing methodology and decreased antibiotic<br />

treatments the ability to detect an association between clinical disease and positive M. bovis<br />

cultures may be altered.<br />

123


5.6 REFERENCES<br />

Biddle, M. K., L. K. Fox, S. J. Evans and C. C. Gay. 2005. Pulsed-field gel electrophoresis<br />

patterns <strong>of</strong> mycoplasma isolates from various body sites in dairy cattle with mycoplasma<br />

mastitis. JAVMA. 227:455-459.<br />

Carroll, J. and N. E. Forsberg. 2007. Influence <strong>of</strong> stress and nutrition on cattle immunity. Vet<br />

Clin Food Anim. 23:105-149.<br />

Caswell, J. L. and M. Archambault. 2007. Mycoplasma bovis pneumonia in cattle. Animal<br />

Health Research Reviews. 8:161-186.<br />

Catry, B., L. Duchateau, VenVan DE, H. Laevens, G. Opsomer, F. Haesebrouck and De Kruif A.<br />

2008. Efficacy <strong>of</strong> metaphylactic florfenicol therapy during natural outbreaks <strong>of</strong> bovine<br />

respiratory disease. J. Vet. Pharmacol. Ther. 31:479-487.<br />

Curtis, C. R., J. M. Scarlett, H. N. Erb and M. E. White. 1988. Path model <strong>of</strong> individual-calf risk<br />

factors for calfhood morbidity and mortality in New York Holstein herds. Prev Vet Med. 6:43-<br />

62.<br />

Dohoo, I. R., S. W. Martin and H. Stryhn. 2003. Veterinary Epidemiologic Research. 1st ed.<br />

AVC Inc, Charlottetown, PEI.<br />

Donovan, G. A., I. R. Dohoo, D. M. Montgomery and F. L. Bennett. 1998. Associations between<br />

passive immunity and morbidity and mortality in dairy heifers in Florida, USA. Prev. Vet. Med.<br />

34:31-46.<br />

Godinho, K. S., A. Rae, G. D. Windsor, N. Tilt, T. G. Rowan and S. J. Sunderland. 2005.<br />

Efficacy <strong>of</strong> tulathromycin in the treatment <strong>of</strong> bovine respiratory disease associated with induced<br />

mycoplasma bovis infections in young dairy calves. Vet. Ther. 6:96-112.<br />

Godinho, K. S., P. Sarasola, E. Renoult, N. Tilt, S. Keane, G. D. Windsor, T. G. Rowan and S. J.<br />

Sunderland. 2007. Use <strong>of</strong> deep nasopharyngeal swabs as a predictive diagnostic method for<br />

natural respiratory infections in calves. Vet Rec. 160:22-25.<br />

Hogan, J. S., R. N. González, R. J. Harmon, S. C. Nickerson, S. P. Oliver, J. W. Pankey and K.<br />

L. Smith. 1999. National Mastitis Council Laboratory Handbook on Bovine Mastitis. National<br />

Mastitis Council, Madison, Wisconsin.<br />

Maunsell, F. P. and G. A. Donovan. 2009. Mycoplasma bovis infections in young calves.<br />

Veterinary Clinics <strong>of</strong> North America: Food Animal Practice. 25:139-177.<br />

Maunsell, F. and G. A. Donovan. 2008. Biosecurity and risk management for dairy replacements.<br />

Veterinary Clinics <strong>of</strong> North America: Food Animal Practice. 24:155-190.<br />

124


McGuirk, S. M. 2008. Disease management <strong>of</strong> dairy calves and heifers. Vet. Clin. Food. Anim.<br />

24:139-153.<br />

Nowakowski, M. A., P. B. Inskeep, J. E. Risk, T. L. Skogerboe, H. A. Benchaoui, T. R. Meinert,<br />

J. Sherington and S. J. Sunderland. 2004. Pharmacokinetics and lung tissue concentrations <strong>of</strong><br />

tulathromycin, a new triamilide antibiotic, in cattle. Vet. Ther. 5:60-74.<br />

Sivula, N. J., T. R. Ames, W. E. Marsh and R. E. Werdin. 1996. Descriptive epidemiology <strong>of</strong><br />

morbidity and mortality in Minnesota dairy heifer calves. Prev. Vet. Med. 27:155-171.<br />

Sockett, D. C., S. A. Jicinsky, T. J. Earleywine, B. L. Miller, B. S. Johnson and J. D. Olson.<br />

2008. Efficacy <strong>of</strong> tulathromycin and oxytetracycline on reducing the incidence <strong>of</strong> otitis media<br />

caused by mycoplasma bovis in preweaned Holstein dairy calves. American Association <strong>of</strong><br />

Bovine Practioners Conference Proceedings, Charlotte, North Carolina. Page 8 .<br />

<strong>Stanton</strong>, A. L., D. F. Kelton, S. J. LeBlanc, S. T. Millman, J. Wormuth, R. T. Dingwell and K. E.<br />

Leslie. 2010. The effect <strong>of</strong> treatment with long-acting antibiotic at postweaning movement on<br />

respiratory disease and on growth in commercial dairy calves. J. Dairy Sci. 93:574-581.<br />

Stipkovits, L., P. H. Ripley, M. Tenk, R. Glávits, T. Molnár and L. Fodor. 2005. The efficacy <strong>of</strong><br />

valnemulin (econor®) in the control <strong>of</strong> disease caused by experimental infection <strong>of</strong> calves with<br />

mycoplasma bovis. Res. Vet. Sci. 78:207-215.<br />

Svensson, C., P. Liberg and J. Hultgren. 2007. Evaluating the efficacy <strong>of</strong> serum haptoglobin<br />

concentration as an indicator <strong>of</strong> respiratory-tract disease in dairy calves. The Veterinary Journal.<br />

174(2):288-94.<br />

Svensson, C., K. Lundborg, U. Emanuelson and S. Olsson. 2003. Morbidity in Swedish dairy<br />

calves from birth to 90 days <strong>of</strong> age and individual calf-level risk factors for infectious diseases.<br />

Prev. Vet. Med. 58:179-197.<br />

Trotz-Williams, L. A., K. E. Leslie and A. S. Peregrine. 2008. Passive immunity in Ontario dairy<br />

calves and investigation <strong>of</strong> its association with calf management practices. J. Dairy Sci. 91:3840-<br />

3849.<br />

Tyler, J. W., D. D. Hancock, S. E. Wiksie, S. L. Holler, J. M. Gay and C. C. Gay. 1998. Use <strong>of</strong><br />

serum protein concentration to predict mortality in mixed-source dairy replacement heifers.<br />

Journal <strong>of</strong> Veterinary Internal Medicine. 12:79-83.<br />

USDA. 2010. Dairy 2007, heifer calf health and management practices on U.S. dairy operations,<br />

2007.USDA-APHIS:VS, CEAH. Fort Collins, CO.#550.0110<br />

USDA. 2009. Dairy 2007, part IV: Reference <strong>of</strong> dairy cattle health and management practices in<br />

the United States, 2007. USDA-APHIS: VS, CEAH. Fort Collins, CO. #N494.0209<br />

125


USDA. 2007. Dairy 2007, part II: Changes in the U.S. dairy cattle industry, 1991–2007.<br />

USDA:APHIS:VS, CEAH. Fort Collins, CO #N517.0208<br />

USDA. 2000. Part III: Health management and biosecurity in U.S. feedlots, 1999.<br />

USDA:APHIS:VS, CEAH. Fort Collins, CO. #N336.1200<br />

Van Donkersgoed, J., C. Ribble, L. G. Boyer and H. Townsend. 1993. Epidemiological study <strong>of</strong><br />

enzootic pneumonia in dairy calves in Saskatchewan. Can J Vet Res. 57:247-254.<br />

Van Donkersgoed, J. 1992. Meta-analysis <strong>of</strong> field trials <strong>of</strong> antimicrobial mass medication for<br />

prophylaxis <strong>of</strong> bovine respiratory disease in feedlot cattle. The Canadian Veterinary Journal.<br />

33(12) p. 786-795.:.<br />

Virtala, A. M. K., G. D. Mechor and Y. T. Grohn. 1996a. The effect <strong>of</strong> calfhood diseases on<br />

growth <strong>of</strong> female dairy calves during the first 3 months <strong>of</strong> life in New York State. Journal <strong>of</strong><br />

Dairy Science. 79:1040-1049.<br />

Virtala, A. M., G. D. Mechor, Y. T. Gröhn, H. N. Erb and E. J. Dubovi. 1996b. Epidemiologic<br />

and pathologic characteristics <strong>of</strong> respiratory tract disease in dairy heifers during the first three<br />

months <strong>of</strong> life. Journal <strong>of</strong> the American Veterinary Medical Association. 208:2035-2042.<br />

Waltner-Toews, D., S. W. Martin, A. H. Meek and I. McMillan. 1986. Dairy calf management,<br />

morbidity and mortality in Ontario Holstein herds. I. the data. Preventive Veterinary Medicine.<br />

4:103-124.<br />

126


Table 5.1: Disease definitions used by farm staff<br />

Disease Definition<br />

Ear Droop<br />

Neonatal Calf<br />

Diarrhea Complex<br />

(Diarrhea)<br />

Bovine Respiratory<br />

Disease (BRD)<br />

Non-Specific Fever<br />

(fever)<br />

Non-Specific<br />

Lameness (lameness)<br />

One or both ears are visibly drooping or head is maintained in a tilted<br />

position, Figure 5.1<br />

Feces are s<strong>of</strong>t and do not hold form<br />

Rectal temperature above 39.5 C plus one or more <strong>of</strong> the following:<br />

elevated respiratory rate, nasal discharge, cough<br />

Dull and listless with a rectal temperature above 39.5C<br />

Abnormal gait with unknown cause<br />

Navel infection Navel is swollen or tender<br />

127


Table 5.2: Incidence and median age at first treatment for all health events for calves at a<br />

commercial heifer raising facility during the first 8 weeks <strong>of</strong> life.<br />

Health Events Incidence (%) Median Age at<br />

first treatment<br />

(days)<br />

Neonatal calf diarrhea 659/788 (84%) 10<br />

complex<br />

Otitis Media 616/788 (78%) 8<br />

Unilateral Ear Droop 32/788 (4%) 8<br />

Bilateral Ear Droop 414/788 (52%) 7<br />

Unspecified Ear Droop 170/788 (22%) 11<br />

Non-specific fever 115/788 (15%) 28<br />

Bovine Respiratory Disease 21/788 (3%) 35<br />

Lameness 16/788 (2%) 17<br />

Navel ill 16/788 (2%) 7<br />

Hernia 5/788 (1%) 19<br />

Bloat 2/788 (< 1%) 17<br />

Failure <strong>of</strong> Passive Transfer 40/778 (5%)<br />

No Disease Treatments 27/788 (3%)<br />

128


Table 5.3: Incidence <strong>of</strong> health events for calves sampled for Mycoplasma at 0, 2 and 4 weeks <strong>of</strong><br />

age.<br />

Health Condition Health conditions<br />

with 1 positive<br />

culture <strong>of</strong><br />

Mycoplasma at 2<br />

or 4 weeks old<br />

(n = 17)<br />

Health conditions<br />

in Mycoplasma<br />

negative calves<br />

(n = 49)<br />

129<br />

P-Values<br />

Diarrhea 1 14/17 (82%) 45/49 (92%) 0.36<br />

Ear Droop 10/17 (59%) 39/49 (78%) 0.11<br />

Fever 2 1/17 (6%) 7/49 (14%) 0.67<br />

BRD 3 1/17 (6%) 1/49 (2%) 0.45<br />

Lameness 4 0/17 (0%) 2/49 (4%) 1.0<br />

Navel ill 1/17 (6%) 2/49 (4%) 1.0<br />

FPT 5 0/17 (0%) 5/49 (10%) 0.32<br />

No health<br />

conditions<br />

2/17 (12%) 3/49 (6%)<br />

0.60<br />

1 Diarrhea: Neonatal Calf Diarrhea Complex<br />

2 Fever: Non-specific Fever<br />

3: BRD: Bovine Respiratory Disease Complex<br />

4 Lameness: Non-specific Lameness<br />

5 FPT: Failure <strong>of</strong> Passive Transfer


Table 5.4: Breakdown <strong>of</strong> the number <strong>of</strong> positive M. Bovis samples collected from calves at<br />

repeated intervals (0, 2, and 4) by source farm and strain type<br />

Farm Sample Types Proportion <strong>of</strong> calves with Strain M.Bovis Freezer<br />

positive Mycoplasma<br />

(strain Death<br />

samples (pos/n sampled) A B C D unknown)<br />

1 Repeated 22% (8/36) 2 - - 1 4 1<br />

2 Repeated 20% (3/15) 1 - 1 - - 1<br />

3 Repeated 38% (3/8) - 1 - - 1 1<br />

4 Repeated 60% (3/5) - 1 1 - - 1<br />

5 Repeated 0% (0/2) - - - - - -<br />

TOTAL 66 3 2 2 1 5 4<br />

130


Table 5.5: Incidence <strong>of</strong> health conditions as reported in the literature for calves under 6 months<br />

<strong>of</strong> age<br />

Health Condition Study<br />

Incidence<br />

Neonatal Calf Diarrhea<br />

Complex<br />

Bovine Respiratory<br />

Disease<br />

Incidence in<br />

Literature<br />

84% 9.8-35%<br />

3%<br />

Non-specific fever 15%<br />

7-39%<br />

Navel ill 2% 1.3-14%<br />

131<br />

References<br />

Waltner-Toews et al., 1986; Curtis et<br />

al., 1988; Van Donkersgoed et al.,<br />

1993; Sivula et al., 1996; Virtala et<br />

al., 1996a; Donovan et al., 1998;<br />

Svensson et al., 2003; USDA, 2010<br />

Waltner-Toews et al., 1986; Van<br />

Donkersgoed et al., 1993; Sivula et<br />

al., 1996; Virtala et al., 1996b;<br />

Donovan et al., 1998; Svensson et al.,<br />

2003; USDA, 2010<br />

Virtala et al., 1996b; Donovan et al.,<br />

1998; Svensson et al., 2003; USDA,<br />

2010; Svensson et al. 2007,<br />

Hernia 1% 15% Virtala et al., 1996a<br />

Failure <strong>of</strong> passive<br />

transfer<br />

5% 19-37%<br />

Mortality 1.9% 3.8-12%<br />

Van Donkersgoed et al., 1993; Virtala<br />

et al., 1996a; Trotz-Williams et al.,<br />

2008; USDA, 2010<br />

Waltner-Toews et al., 1986; Curtis et<br />

al., 1988; Sivula et al., 1996; USDA,<br />

2010


Figure 5. 5.1: 1: Images taken from farm farm farm farm farm farm prior prior prior prior prior prior to to to to to to the the start <strong>of</strong> the the the the the the the the trial trial trial trial trial trial and used as g<br />

g<br />

g<br />

g<br />

g<br />

g<br />

g<br />

guidelines uidelines to<br />

to<br />

to<br />

identify ear ear ear ear ear ear droop. droop. droop. droop. droop. droop. droop. droop. Shown: Shown: Shown: Shown: Shown: Shown: Shown: Shown: from from from from from from from left left left left left left to to to to to to to to right: right: right: right: right: right: right: right: right: right: right: Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral ear ear ear ear ear ear ear ear droop droop droop droop and and and and and and and and ear ear ear ear ear droop droop droop droop droop droop with with with with with with with abnormal<br />

abnormal<br />

abnormal<br />

abnormal<br />

abnormal<br />

abnormal<br />

abnormal<br />

abnormal<br />

head position<br />

132


CHAPTER 6: THE EFFECT OF TREATMENT WITH LONG-ACTING<br />

ANTIBIOTIC AT POSTWEANING MOVEMENT ON RESPIRATORY DISEASE<br />

AND ON GROWTH IN COMMERCIAL DAIRY CALVES<br />

A.L. <strong>Stanton</strong> 1* , D.F. Kelton 1 , S.J. LeBlanc 1 , S.T. Millman 2 , J. Wormuth 3 ,<br />

R.T. Dingwell 1 and K.E. Leslie 1<br />

1 Department <strong>of</strong> Population Medicine, <strong>University</strong> <strong>of</strong> <strong>Guelph</strong>, <strong>Guelph</strong>, N1G 2W1 Ontario, Canada<br />

2 Veterinary Diagnostic and Production Animal Medicine/Biomedical Sciences, College <strong>of</strong><br />

Veterinary Medicine, Iowa State <strong>University</strong>, Ames 50011<br />

3 CY Heifer Farm, Elba, NY 14058<br />

Key words: (antibiotic; weaning; respiratory disease; heifer)<br />

* Corresponding author.<br />

Mailing Address: Department <strong>of</strong> Population Medicine, Ontario Veterinary College, <strong>University</strong> <strong>of</strong><br />

<strong>Guelph</strong>, <strong>Guelph</strong>, Ontario, Canada N1G 2W1. (519) 824-4120 ext. 52995.<br />

Reprinted with permission from:<br />

Journal <strong>of</strong> Dairy Science. 2010. 93 (2): 574-581.<br />

133


6.1 ABSTRACT<br />

Bovine respiratory disease (BRD) is a major concern when raising replacement heifers<br />

because <strong>of</strong> the high incidence and long-term effects <strong>of</strong> this disease, such as decreased growth and<br />

increased time to first calving. The objective <strong>of</strong> this study was to determine the effect <strong>of</strong><br />

tulathromycin (TUL) treatment at postweaning movement on the incidence <strong>of</strong> BRD in dairy<br />

replacement heifers. A total <strong>of</strong> 1,395 heifers were enrolled between November 2006 and June<br />

2007 at a commercial heifer-raising facility. Calves were randomly assigned either to treatment<br />

with TUL or to a positive control group treated with oxytetracycline (TET). Calves treated with<br />

TUL were 0.5 times (95% CI: 0.4 to 0.7) less likely to be treated for BRD in the 60 d following<br />

enrollment than calves treated with TET. For calves that had no history <strong>of</strong> BRD in the pre-<br />

enrollment period, TET calves weighed 4.9 ± 0.5 kg less than TUL calves after 6 wk in group<br />

housing. If calves were treated for BRD in the pre-enrollment period, there was no treatment<br />

effect on growth. Calves with clinical BRD in the 60 d following movement weighed<br />

7.9 ± 0.6 kg less than calves without BRD after 6 wk in group housing. Treatment with TUL at<br />

the time <strong>of</strong> movement to group housing had a beneficial effect on the health and performance<br />

through the prevention <strong>of</strong> BRD in dairy calves with no prior history <strong>of</strong> the disease. Moreover,<br />

BRD after movement to group housing after weaning had a significant effect on the growth <strong>of</strong><br />

dairy calves.<br />

134


6.2 INTRODUCTION<br />

Dairy replacement heifers undergo numerous challenges from birth until they enter the<br />

lactating herd, with many <strong>of</strong> these challenges coinciding with management changes. In the North<br />

American dairy industry, preweaned dairy heifers are raised primarily in individual housing and<br />

after weaning are primarily housed in groups (USDA, 2007a). Potential stressors involved in this<br />

move from individual to group housing include social, environmental, transportation, and<br />

nutritional changes. The environmental stressors include increased pathogen exposure and<br />

temperature changes. If calves are required to mount a prolonged response to these multiple<br />

stressors, suppression <strong>of</strong> their immune response and disease outbreaks can occur (Carroll and<br />

Forsberg, 2007). Bovine respiratory disease (BRD) is the primary disease to which calves are<br />

susceptible at this time.<br />

In cattle, BRD is a complex <strong>of</strong> disease processes most commonly resulting from infection<br />

with various microorganisms, including Pasteurella multocida, Mannheimia hemolytica, and<br />

Mycoplasma spp. The development <strong>of</strong> BRD is typically initiated by an environmental stressor<br />

and viral infection, which weakens the resistance mechanisms <strong>of</strong> the lungs and allows bacterial<br />

colonization <strong>of</strong> the lung (Yates, 1982). In an Ontario study, 15% <strong>of</strong> calves were treated for<br />

pneumonia before weaning (Waltner-Toews et al., 1986a). In a Minnesota study, the case fatality<br />

rate for BRD was about 9% (Sivula et al., 1996). The direct costs <strong>of</strong> treatment and prevention <strong>of</strong><br />

this disease have been estimated at US$14.71 per calf per year in milk-fed calves, as well as<br />

$1.95 per calf per year in replacement heifers and $9.08 per cow per year over the entire herd<br />

(Kaneene and Scott Hurd, 1990; Miller and Dorn, 1990). These studies were not able to take into<br />

account the indirect costs <strong>of</strong> this disease, which include an increased risk <strong>of</strong> culling or death<br />

135


efore entry into the milking herd or the losses associated with delayed entry into the milking<br />

herd (Waltner-Toews et al., 1986b). Important control measures include vaccinating cows<br />

against the causal organisms, ensuring excellent colostrum management, and minimizing acute<br />

and chronic stressors such as poor ventilation, mixing <strong>of</strong> groups, overcrowding, and poor<br />

nutrition (Kahn, 2005). However, at weaning or first movement, some <strong>of</strong> these stressors cannot<br />

be avoided. As such, calves are placed at increased risk <strong>of</strong> developing BRD and additional steps<br />

may be necessary to prevent BRD.<br />

In 1999, 27% <strong>of</strong> small beef feedlots and 81% <strong>of</strong> large feedlots in the United States were<br />

administering an injectable antimicrobial metaphylactically to feedlot cattle for the prevention <strong>of</strong><br />

respiratory disease (USDA, 2000). A meta-analysis <strong>of</strong> field trials indicated that for feedlot<br />

calves, the administration <strong>of</strong> oxytetracycline or tilmicosin on the day <strong>of</strong> arrival at the feedlot<br />

consistently reduced morbidity rates attributable to BRD (Van Donkersgoed et al., 1993). The<br />

movement <strong>of</strong> beef cattle into a feedlot is comparable to the movement <strong>of</strong> dairy calves from<br />

individual to group housing in that they experience similar potential stressors, including<br />

nutritional and social changes, transportation, and increased disease exposure. However, the<br />

management systems and the age and size <strong>of</strong> the animals are different between beef and dairy<br />

calves. As such, research into the use <strong>of</strong> injectable antimicrobials at the time <strong>of</strong> calf movement to<br />

prevent BRD in weaned dairy calves is necessary to determine the efficacy <strong>of</strong> this approach in<br />

the context <strong>of</strong> a dairy replacement heifer management system.<br />

The objective <strong>of</strong> this randomized blinded clinical trial was to compare the effects <strong>of</strong><br />

tulathromycin under commercial conditions on calf health and performance to an existing<br />

oxytetracycline treatment. Tulathromycin (Draxxin, Pfizer Animal Health Group, New York,<br />

136


NY) is a tiamilide compound, a subclass <strong>of</strong> macrolide long-acting antimicrobials. It is<br />

distinguished by maintaining therapeutic concentrations against pathogens commonly associated<br />

with BRD for up to 10 d (Nowakowski et al., 2004). Tulathromycin is approved for treatment<br />

and control <strong>of</strong> BRD in beef cattle and nonlactating dairy cattle. Tulathromycin has been shown to<br />

be effective in the treatment <strong>of</strong> BRD in 3- to 9-wk-old calves challenged with Mycoplasma bovis<br />

(Godinho et al., 2005, 2007). Oxytetracycline is a broad-spectrum tetracycline antibiotic that<br />

maintains therapeutic concentrations against pathogens associated with BRD for 3 d (Biomycin<br />

200, Boehringer Ingelheim Vetmedica Inc., St. Joseph, MO). Oxytetracycline is approved for use<br />

in beef and dairy cattle. The hypothesis for this study was that administration <strong>of</strong> tulathromycin at<br />

a time <strong>of</strong> high risk <strong>of</strong> BRD would reduce the pathogen challenge and the risk <strong>of</strong> clinical<br />

respiratory disease in the weeks following a postweaning movement and housing change in dairy<br />

calves.<br />

6.3 MATERIALS AND METHODS<br />

6.3.1. Heifer Management and Housing<br />

This study was conducted at a commercial contract heifer raising facility in New York<br />

State. Enrollment in the study occurred between November 2006 and June 2007, with a total <strong>of</strong><br />

1,395 heifers being enrolled.<br />

Heifer calves arrived at the farm from 7 commercial dairy farms at 1 to 5 d <strong>of</strong> age. Upon<br />

arrival at the facility, each calf was processed according to a standard farm protocol and was then<br />

moved to individual housing in a naturally ventilated nursery barn or an outdoor hutch. The farm<br />

protocol included the identification <strong>of</strong> any congenital abnormalities, measurement <strong>of</strong> serum total<br />

protein, tail docking, and the collection <strong>of</strong> a tissue sample for the detection <strong>of</strong> persistent infection<br />

137


with bovine viral diarrhea (BVD). Over the course <strong>of</strong> this study, no animal tested positive for<br />

BVD. Calves were vaccinated at 2 and 4 wk <strong>of</strong> age with an intramuscular modified live vaccine<br />

against infectious bovine rhinotracheitis, BVD type 1 and type 2, bovine respiratory syncytial<br />

virus, and parainfluenza-3 virus (Vista-5, Intervet Schering-Plough Animal Health, Boxmeer, the<br />

Netherlands).<br />

Preweaned calves were fed 4 L <strong>of</strong> milk replacer (Excel Calf Milk Replacer 26/18, Grober<br />

Nutrition, Cambridge, Ontario, Canada) twice daily and had access to ad libitum calf starter and<br />

water. Weaning from milk replacer was initiated at 5 wk and completed at 6 wk <strong>of</strong> age. Calves<br />

remained in their original individual housing for approximately 2 wk after weaning, at which<br />

time they were moved to group housing in the weaning barn and enrolled in the trial. Calves that<br />

were moved to group housing after 80 d <strong>of</strong> age were excluded from this trial. A schematic<br />

representation <strong>of</strong> the trial methods and events is shown in Figure 6.1.<br />

For the first 6 wk <strong>of</strong> the study, calves were housed in the weaning barn in groups <strong>of</strong> 8<br />

to10 calves. For the initial 3 to 4 wk in the weaning barn, calves received dry hay and pelleted<br />

heifer grain ad libitum. Calves were then transitioned onto a TMR. After approximately 6 wk in<br />

the weaning barn, calves were moved to the transition barn and housed in groups <strong>of</strong><br />

approximately 50 calves.<br />

6.3.2 Study Treatments<br />

Before the start <strong>of</strong> the study, the farm had a pre-existing treatment protocol at movement.<br />

As such, the current study was designed as a positive control trial, with calves being assigned to<br />

treatment with tulathromycin (TUL) or the previously established oxytetracycline protocol<br />

(TET). Each calf was assigned to treatment group using a systematic random assignment method<br />

138


ased on the unique identification number assigned to each calf on arrival, determined by their<br />

order <strong>of</strong> arrival at the farm. All TUL calves were injected with 2.0 mL <strong>of</strong> tulathromycin<br />

(Draxxin, Pfizer Animal Health Group) by subcutaneous injection on the day <strong>of</strong> movement to<br />

group housing. Draxxin is a semi-synthetic macrolide antibiotic, containing 100 mg <strong>of</strong><br />

tulathromycin per milliliter. All TET calves received 5.0 mL <strong>of</strong> oxytetracycline (Biomycin 200,<br />

Boehringer Ingelheim, St. Joseph, MO) intramuscularly on the day <strong>of</strong> movement to group<br />

housing. Biomycin 200 is a tetracycline antibiotic containing 200 mg <strong>of</strong> oxytetracycline per<br />

milliliter. The standard dose for TUL was determined based on the weight <strong>of</strong> the upper quartile<br />

<strong>of</strong> calves (78 kg) at enrollment. The dose <strong>of</strong> TET was part <strong>of</strong> the established weaning<br />

management protocol on the farm and was two-thirds <strong>of</strong> a full label dose. During the study, no<br />

antibiotics were administered in the feed or water. The 2 trial treatments were very different in<br />

appearance and volume used; therefore, the farm staff members responsible for administration <strong>of</strong><br />

treatment at the time <strong>of</strong> enrollment were not blinded to treatment assignment. However, the<br />

decision as to which drug to administer was predetermined. As such, randomization to treatment<br />

group was not influenced by preconceived bias <strong>of</strong> the farm staff. Outcomes, such as detection <strong>of</strong><br />

BRD and follow-up weight measurements, were assessed by a different set <strong>of</strong> staff members than<br />

those administering the treatment at enrollment, who were blind to treatment group allocation.<br />

The distribution <strong>of</strong> calves in the TET and TUL groups within a pen <strong>of</strong> 10 calves in the<br />

weaning barn was random, based upon the order in which calves were processed. Pens were<br />

open-sided. Thus, nose-to-nose and oral contact was possible between pens.<br />

139


6.3.3 Outcome Measures<br />

All calves were monitored daily in the nursery, weaning, and transition barns by trained<br />

barn staff for the occurrence <strong>of</strong> disease, using consistent case definitions. All disease events and<br />

treatments were recorded in the farm management database (DairyComp 305, Valley<br />

Agricultural S<strong>of</strong>tware Inc., Tulare, CA). The farm staff recorded fever and respiratory disease as<br />

2 separate events. Fever was defined as an animal that was dull, <strong>of</strong>f-feed, listless, with a rectal<br />

temperature > 39.5°C and normal respiratory rate with no nasal discharge; respiratory disease<br />

was defined as the animal being dull, listless, with elevated respiratory rate, nasal discharge, or<br />

both. However, for the purpose <strong>of</strong> this trial, we combined these 2 diseases into one category<br />

called BRD. This approach was based on the likelihood that these 2 events are the same disease<br />

at different stages or represent animals with a slightly different clinical presentation <strong>of</strong> the same<br />

underlying condition. All calves in individual housing were treated for disease in accordance<br />

with the previously established treatment protocols <strong>of</strong> the farm. Calves were not eligible for<br />

BRD treatment for 5 d after the experimental treatment. Group-housed animals identified with<br />

BRD were treated with cefti<strong>of</strong>ur crystalline suspension (Excede Sterile Suspension, Pfizer<br />

Animal Health Group) administered at the base <strong>of</strong> the ear, according to label instructions.<br />

All calves were weighed upon arrival at the farm, upon enrollment in the study<br />

(approximately 8 wk <strong>of</strong> age), and at 6 wk after enrollment. Height at the withers (to the nearest<br />

inch) was measured at the same times as weight.<br />

Farm staff performed a gross necropsy for those calves with signs <strong>of</strong> BRD that died<br />

during the first 6 wk post-enrollment. To complete this gross postmortem examination, specific<br />

digital images <strong>of</strong> the heart, lung, intestines, and trachea were taken. These images, along with a<br />

140


case history, were sent electronically to a pathologist at the <strong>University</strong> <strong>of</strong> <strong>Guelph</strong> (Ontario,<br />

Canada) for a morphological diagnosis. Calves that died as a result <strong>of</strong> a known problem such as<br />

physical injury or were culled from the herd did not have a necropsy performed. If necropsy<br />

results were not available, staff-designated clinical findings and diagnosis were used as the most<br />

likely cause <strong>of</strong> death.<br />

6.3.4 Statistical Analysis<br />

All analyses were performed using SAS statistical s<strong>of</strong>tware (SAS Institute Inc., Cary,<br />

NC). Descriptive statistics and residual diagnostics were performed using Proc Freq and Proc<br />

Univariate procedures. Regression analyses were conducted using a mixed linear model (Proc<br />

Mixed) for height and weight at exit from the weaning barn, and a generalized linear mixed<br />

model with a logit-transform (Glimmix) was used to analyze the probability <strong>of</strong> BRD in the first<br />

60 d post-enrollment. The effects <strong>of</strong> treatment and BRD on weight and height were modeled<br />

separately. The models for weight controlled for enrollment weight and age at exit measurement.<br />

The models for height controlled for enrollment height and age at exit measurement. Clustering<br />

by weekly enrollment cohort and farm <strong>of</strong> origin were included in all models as random effects.<br />

Disease variables considered for analysis are described in Table 6.1. In preliminary analysis, pre-<br />

enrollment disease events were divided into 2 time periods, with the disease occurring either 2<br />

wk before enrollment (between weaning and enrollment) or from birth to 2 wk before enrollment<br />

(weaning). Variables that were significant were found to be <strong>of</strong> similar magnitude and direction in<br />

the model; as such, these time periods were combined into one pre-enrollment category. All<br />

variables that were potentially associated (P < 0.2) with the outcomes in univariate analyses were<br />

included in the <strong>final</strong> model and removed by backward step-wise elimination based on partial F-<br />

141


tests, or multiple partial F-tests, with a P-value < 0.05. Confounding was assessed if the<br />

exclusion <strong>of</strong> a variable changed the remaining parameter estimates by 30%. Disease variables<br />

were tested for potential interactions with treatment. All data were assessed for the assumption <strong>of</strong><br />

normality. Weight and height models did not include calves that died before their <strong>final</strong><br />

measurement.<br />

6.4 RESULTS<br />

Three calves were removed from all analysis. One TUL calf was diagnosed and treated<br />

for BRD at 1 d post-enrollment, which violated the protocol. One TET calf and one TUL calf<br />

were removed from analysis because they were more than 80 d <strong>of</strong> age at the time <strong>of</strong> enrollment<br />

on the trial. In total, 1,392 animals were included in the analysis <strong>of</strong> <strong>final</strong> outcomes.<br />

Calves arrived at the trial facility at 3 ± 2.1 (mean ± SD) d <strong>of</strong> age and were 56 ± 7 d <strong>of</strong><br />

age at enrollment. At enrollment, calves weighed 71 ± 11 kg and were 90 ± 4 cm at the withers.<br />

Each week, 40 ± 11 calves were enrolled in the trial. Pre-enrollment, 63.7% (887/1,392) <strong>of</strong><br />

calves were treated for scours, 59.1% (822/1,392) were treated for otitis media, 14.3%<br />

(199/1,392) were treated for BRD, and 1.7% (24/1,392) were treated for lameness. Before<br />

enrollment, 73.3% (147/199) <strong>of</strong> calves treated for pre-enrollment BRD were also treated for<br />

otitis media. Of the calves treated for both otitis media and BRD pre-enrollment, 74.0%<br />

(108/147) were treated for otitis media before BRD. Age, weight, and height at enrollment, and<br />

pre-enrollment diseases were not significantly different between TET and TUL calves.<br />

6.4.1 Incidence <strong>of</strong> BRD in the 60 d Post-Movement<br />

Risk <strong>of</strong> treatment for BRD for the first 60 d following enrollment was evaluated for 1,392<br />

calves. In the 6 wk after enrollment, the frequency <strong>of</strong> treatment for BRD was 22.4% (156/695) in<br />

142


the TET calves and 13.2% (92/697) in the TUL calves. The mean time <strong>of</strong> treatment <strong>of</strong> BRD was<br />

24.7 ± 13.4 d after enrollment and was similar (P = 0.58) among calves in the TET and TUL<br />

groups. The model <strong>of</strong> the outcome BRD included 3 significant fixed effects: treatment, pre-<br />

enrollment lameness, and pre-enrollment average daily gain (Table 6.2). There were no<br />

significant interaction terms. Controlling for the other variables in the model, TET calves were<br />

2.0 times (95% CI: 1.5 to 2.6) more likely to be treated for BRD in the 60 d following enrollment<br />

than were TUL calves (P < 0.0001). There was a tendency for calves treated for preweaning<br />

lameness to be 2.3 times (95% CI: 0.9 to 6.0) more likely to be treated for BRD (P = 0.07). As<br />

pre-enrollment average daily gain increased, risk <strong>of</strong> BRD decreased (P < 0.005). For example, a<br />

calf with a pre-enrollment average daily gain <strong>of</strong> 0.5 kg/d was 1.5 times more likely to be treated<br />

for BRD than a calf with a pre-enrollment average daily gain <strong>of</strong> 0.75 kg/d.<br />

6.4.2 Growth in Postweaned Dairy Calves<br />

Twenty-one calves died before being weighed 6 wk after enrollment. Therefore 1,371<br />

calves had exit weights and heights and were included in this analysis. The data were analyzed to<br />

determine the effect <strong>of</strong> treatment and the effect <strong>of</strong> BRD on the weight and height <strong>of</strong> calves upon<br />

exiting the weaning barn, 6 wk after enrollment. Overall, calves gained 37.3 ± 11.8 kg during the<br />

6 wk post-enrollment and grew on average 8 ± 0.1 cm.<br />

The effect <strong>of</strong> treatment on weight after exiting the weaning barn had a significant<br />

interaction with pre-enrollment BRD. If calves were not treated for BRD pre-enrollment, the<br />

TUL treatment increased the exit weight <strong>of</strong> calves by 4.9 ± 0.5 kg relative to TET calves<br />

(P < 0.001; Table 6.3). If calves were treated for BRD in the pre-enrollment period, treatment<br />

had no effect on BW (P = 0.38).<br />

143


The effect <strong>of</strong> treatment on height after exiting the weaning barn had significant<br />

interactions with pre-enrollment scours and pre-enrollment BRD. Among calves with no history<br />

<strong>of</strong> BRD, TUL increased the exit height <strong>of</strong> calves by 0.8 ± 0.1 cm (P < 0.001) and, among calves<br />

with no history <strong>of</strong> preweaning scours, TUL increased the exit height <strong>of</strong> by 0.7 ± 0.3 cm<br />

(P < 0.05) compared with TET calves (Table 6.4). Treatment had no significant effect on calves<br />

with a history <strong>of</strong> pre-enrollment scours or pre-enrollment BRD.<br />

Bovine respiratory disease in the pre-enrollment or post-enrollment period affected the<br />

weight <strong>of</strong> calves when they exited the weaning barn. Accounting for weight at enrollment, age at<br />

measurement, and random effects <strong>of</strong> source farm and enrollment cohort, pre-enrollment BRD<br />

decreased the exit weight by 2.9 ± 0.7 kg (P < 0.001) and post-enrollment BRD decreased exit<br />

weight by 7.9 ± 0.6 kg (P < 0.001).<br />

Bovine respiratory disease in the pre-enrollment or post-enrollment period decreased the<br />

height <strong>of</strong> calves when they exited the weaning barn. Accounting for height at enrollment, age at<br />

measurement, and random effects <strong>of</strong> source farm and enrollment cohort, pre-enrollment BRD<br />

decreased the exit height by 0.6 ± 0.2 (P < 0.005) and post-enrollment BRD decreased the exit<br />

height by 1.2 ± 0.2 (P < 0.001).<br />

6.4.3 Death Loss<br />

Calves were followed for 60 d following enrollment. During this time, 27 calves (1.9%)<br />

were removed from the herd because <strong>of</strong> mortality or culling. Assignment to TUL or TET groups<br />

was not associated with mortality, in that 11 and 16 calves from the TUL and TET groups were<br />

removed, respectively (P = 0.33). The cause <strong>of</strong> death for these calves was most frequently<br />

attributed to BRD (Table 6.5). Ten calves were examined with a digital necropsy because <strong>of</strong><br />

144


suspected respiratory disease or unknown cause <strong>of</strong> death. Of these cases, BRD was determined to<br />

be the most likely cause <strong>of</strong> death in 9 <strong>of</strong> the 10 cases, with one case <strong>of</strong> probable enteritis.<br />

Morphologically, bronchopneumonia was present in 8 out <strong>of</strong> 9 cases, and there was one case <strong>of</strong><br />

probable septicemia due to fibrinous pleuritis. Of the 8 cases <strong>of</strong> bronchopneumonia, Mycoplasma<br />

bovis was the most likely etiologic agent in 3 <strong>of</strong> the cases; the remaining 5 cases were diagnosed<br />

as having a nonspecific bacterial cause based on the gross appearance <strong>of</strong> the lesions. Other than<br />

BRD, deaths were attributed to injury, joint infection, or bloat. Other calves were sold because <strong>of</strong><br />

poor growth and conformation. Death or removal from the farm occurred at 38.7 ± 2.7 d after<br />

enrollment and did not differ by treatment (P = 0.51).<br />

6.5 DISCUSSION<br />

This study evaluated a novel approach to reducing the incidence and effect <strong>of</strong> BRD in<br />

dairy calves, examining the effect <strong>of</strong> a long-acting prophylactic treatment relative to a shorter<br />

acting prophylactic treatment, which is a practice that has been examined primarily in beef cattle<br />

on arrival in feedlots. The time period immediately following movement from individual to<br />

group housing is particularly critical for dairy calves. At this time, there is an increased exposure<br />

to pathogens, an immature immune system, and many social, environmental, and dietary<br />

changes. All <strong>of</strong> these stressors put dairy calves at increased risk for the development <strong>of</strong> BRD<br />

during this time period, similar to beef calves arriving in feedlots.<br />

The objective <strong>of</strong> this study was to compare the effect <strong>of</strong> administering a long-acting<br />

antibiotic, tulathromycin, at the time <strong>of</strong> movement to group housing on the incidence <strong>of</strong> BRD<br />

and effects on growth in dairy calves relative to a shorter acting antibiotic, oxytetracycline. The<br />

study was designed as a positive control study. It should be noted that the positive control used in<br />

145


this study involved continuing with an existing practice at the study farm <strong>of</strong> administering<br />

oxytetracycline around the time <strong>of</strong> weaning to all calves. The dose <strong>of</strong> oxytetracycline used was<br />

derived from the previously established program and was continued into the current study.<br />

Calculation <strong>of</strong> the dosage <strong>of</strong> oxytetracycline that would be appropriate based on the weights <strong>of</strong><br />

calves in this study indicates that the dose <strong>of</strong> TET was lower than is considered to be optimal. As<br />

such, the underdosing <strong>of</strong> oxytetracycline may not represent the effects that would be expected<br />

with a full dose.<br />

The National Animal Health Monitoring and Surveillance (NAHMS) reports on dairy<br />

heifers found that 5% <strong>of</strong> postweaned heifers in the United States were treated for BRD in 2002,<br />

and 6% were treated in 2007 (USDA, 2005, 2007b). However, the NAHMS study defined<br />

postweaned heifers as including all animals from weaning to first calving. The wide age<br />

variation encompassed by this definition would include many older animals that were at low risk<br />

<strong>of</strong> BRD. Preweaned calves have reported rates <strong>of</strong> disease ranging from 0.1 cases/100 d in<br />

Minnesota (Sivula et al., 1996) to 39% <strong>of</strong> calves in Saskatchewan (Van Donkersgoed et al.,<br />

1993). A study <strong>of</strong> herds in New York found that 25% <strong>of</strong> calves in the first 3 mo <strong>of</strong> life were<br />

treated for BRD (Virtala et al., 1996). This substantial variation in the reported rates <strong>of</strong> disease<br />

probably results from a combination <strong>of</strong> the differences in the diagnostic criteria used, age <strong>of</strong> the<br />

calves, and management conditions. In large beef feedlots, 15.5% <strong>of</strong> cattle are treated for BRD,<br />

and on small beef feedlots, 8.7% <strong>of</strong> cattle are treated for BRD (USDA, 2000). During the course<br />

<strong>of</strong> this study, 22% <strong>of</strong> calves in the TET group developed BRD, which is consistent with the<br />

above observational studies <strong>of</strong> dairy calves but is higher than what is found in beef feedlots. The<br />

146


elevated incidence <strong>of</strong> respiratory disease in postweaned dairy calves relative to feedlot cattle<br />

stresses the importance <strong>of</strong> finding effective techniques to manage this disease.<br />

The results <strong>of</strong> this study indicate that calves in the TUL group were half as likely to be<br />

treated for BRD as calves in the TET group during the 60 d following movement. However,<br />

treatment did not affect mortality in the 60 d following movement. This finding may be a result<br />

<strong>of</strong> several factors. First, this was a positive control study so both groups <strong>of</strong> calves were receiving<br />

antibiotic therapy. Second, the overall mortality rate was low and if a difference between the 2<br />

treatment groups did exist, the sample size would not be sufficient to detect it. Finally, given the<br />

long-term increased risk <strong>of</strong> death associated with clinical BRD documented by Waltner-Toews et<br />

al., (1986b), it is also possible that any difference in mortality risk between groups would not be<br />

seen until animals are older.<br />

The failure <strong>of</strong> calves treated for pre-enrollment BRD to respond to TUL with improved<br />

growth indicates that this treatment does not compensate for an earlier occurrence <strong>of</strong> the disease.<br />

This may be because <strong>of</strong> the effect that BRD has already had on the calf's future health and<br />

performance. Despite its high incidence, otitis media was not a significant risk factor for post-<br />

enrollment BRD or decreased growth in this population <strong>of</strong> calves. There may be several reasons<br />

for this, including a program for prompt detection and treatment <strong>of</strong> otitis media on this farm that<br />

may have limited the effect <strong>of</strong> this disease. In addition, given that most calves treated for pre-<br />

enrollment BRD were also treated for otitis media, the effect <strong>of</strong> otitis media may have been<br />

accounted for when pre-enrollment BRD was included in the model. In this study, the<br />

implementation <strong>of</strong> the prophylactic therapy program at the time <strong>of</strong> movement and grouping is <strong>of</strong><br />

the greatest benefit for calves that do not have a history <strong>of</strong> BRD before enrollment. Therefore, it<br />

147


is important to emphasize that the health <strong>of</strong> pre-enrollment calves should be closely monitored to<br />

minimize this disease in the pre-enrollment period as well.<br />

Among calves that were not treated for BRD in the pre-enrollment period, at 6 wk post-<br />

enrollment, calves in the TUL group weighed more and were taller compared with calves in the<br />

TET group. This further supports the hypothesis that TUL calves had a decreased incidence <strong>of</strong><br />

BRD. A previous study documented a decrease in average daily gain <strong>of</strong> 0.14 g/d in calves treated<br />

for BRD during their third month <strong>of</strong> life and a decreased height <strong>of</strong> 0.2 cm for each week <strong>of</strong> BRD<br />

(Virtala et al., 1996). When this value is converted to a 6-wk time period, it becomes a loss in<br />

potential weight <strong>of</strong> 5.9 kg and potential height <strong>of</strong> 1.2 cm. In this study, calves with BRD had<br />

lower weight (7.8 ± 0.5 kg) and height (1.3 ± 0.2 cm). Calves in the TUL group showed<br />

improved weight gain <strong>of</strong> 4.8 ± 1.2 kg and height <strong>of</strong> 1.5 ± 0.3 cm compared with the TET group.<br />

The results <strong>of</strong> the current study indicate that BRD has a highly detrimental effect on the growth<br />

<strong>of</strong> dairy calves and that the administration <strong>of</strong> TUL at the start <strong>of</strong> this high-risk period can<br />

decrease the incidence <strong>of</strong> the disease and mitigate the negative outcomes associated with it.<br />

Knowledge <strong>of</strong> the benefits and limitations <strong>of</strong> a strategic antimicrobial intervention would be<br />

beneficial when management efforts are unable to overcome the risk <strong>of</strong> BRD. In addition, this<br />

approach would be beneficial to protect calves that are at high risk <strong>of</strong> BRD while contributing<br />

factors such as housing and pre-weaning disease are being addressed by farm management.<br />

In conclusion, the use <strong>of</strong> TUL at the time <strong>of</strong> movement to group housing had a beneficial<br />

effect on the health and performance <strong>of</strong> dairy calves compared with treatment with TET,<br />

primarily through the prevention <strong>of</strong> respiratory disease in the period immediately following<br />

movement from individual to group housing.<br />

148


6.6. ACKNOWLEDGMENTS<br />

The authors acknowledge Jeff Caswell (Department <strong>of</strong> Pathobiology, <strong>University</strong> <strong>of</strong><br />

<strong>Guelph</strong>, Ontario, Canada) and the management and staff at CY Heifer Farm (Elba, NY) for their<br />

invaluable contribution to this research. Funding was provided by Pfizer Animal Health (New<br />

York, NY) and the National Science and Engineering Research Council <strong>of</strong> Canada (Ottawa,<br />

Ontario).<br />

149


6.7 REFERENCES<br />

J. Carroll and N.E. Forsberg, Influence <strong>of</strong> stress and nutrition on cattle immunity, Vet. Clin.<br />

North Am. Food Anim. 23 (2007), pp. 105–149.<br />

K.S. Godinho, A. Rae, G.D. Windsor, N. Tilt, T.G. Rowan and S.J. Sunderland, Efficacy <strong>of</strong><br />

tulathromycin in the treatment <strong>of</strong> bovine respiratory disease associated with induced<br />

Mycoplasma bovis infections in young dairy calves, Vet. Ther. 6 (2005), pp. 96–112.<br />

K.S. Godinho, P. Sarasola, E. Renoult, N. Tilt, S. Keane, G.D. Windsor, T.G. Rowan and S.J.<br />

Sunderland, Use <strong>of</strong> deep nasopharyngeal swabs as a predictive diagnostic method for natural<br />

respiratory infections in calves, Vet. Rec. 160 (2007), pp. 22–25.<br />

C.M Kahn, Merck Veterinary Manual (9th ed.), Merck, Whitehouse Station, NJ (2005).<br />

J.B. Kaneene and H. Scott Hurd, The National Animal Health Monitoring System in Michigan.<br />

III. Cost estimates <strong>of</strong> selected dairy cattle diseases, Prev. Vet. Med. 8 (1990), pp. 127–140.<br />

G.Y. Miller and C.R. Dorn, Costs <strong>of</strong> dairy cattle diseases to producers in Ohio, Prev. Vet. Med. 8<br />

(1990), pp. 171–182.<br />

M.A. Nowakowski, P.B. Inskeep, J.E. Risk, T.L. Skogerboe, H.A. Benchaoui, T.R. Meinert, J.<br />

Sherington and S.J. Sunderland, Pharmacokinetics and lung tissue concentrations <strong>of</strong><br />

tulathromycin, a new triamilide antibiotic, in cattle, Vet. Ther. 5 (2004), pp. 60–74.<br />

N.J. Sivula, T.R. Ames, W.E. Marsh and R.E. Werdin, Descriptive epidemiology <strong>of</strong> morbidity<br />

and mortality in Minnesota dairy heifer calves, Prev. Vet. Med. 27 (1996), pp. 155–171.<br />

USDA. 2000. Part III: Health management and biosecurity in U.S. feedlots, 1999. #N336.1200.<br />

USDA:APHIS:VS, CEAH, National Animal Health Monitoring System, Fort Collins, CO.<br />

USDA. 2005. Part IV: Antimcrobial Use on U.S. Dairy Operations, 2002. N430.0905.<br />

USDA:APHIS:VS:CEAH, National Animal Health Monitoring System, Fort Collins, CO.<br />

USDA. 2007a. Dairy 2007, Part I: Reference <strong>of</strong> Dairy Cattle Health and Management Practices<br />

in the United States, 2007. #N480.1007. USDA-APHIS-VS, CEAH, National Animal Health<br />

Monitoring System, Fort Collins, CO.<br />

USDA. 2007b. Dairy 2007, Part III: Reference <strong>of</strong> Dairy Cattle Health and Management Practices<br />

in the United States, 2007. #N480.1007. USDA-APHIS-VS, CEAH, National Animal Health<br />

Monitoring System, Fort Collins, CO.<br />

J. Van Donkersgoed, C. Ribble, L.G. Boyer and H. Townsend, Epidemiological study <strong>of</strong><br />

enzootic pneumonia in dairy calves in Saskatchewan, Can. J. Vet. Res. 57 (1993), pp. 247–254.<br />

150


A.M.K. Virtala, G.D. Mechor and Y.T. Grohn, The effect <strong>of</strong> calfhood diseases on growth <strong>of</strong><br />

female dairy calves during the first 3 months <strong>of</strong> life in New York State, J. Dairy Sci. 79 (1996),<br />

pp. 1040–1049.<br />

D. Waltner-Toews, S.W. Martin and A.H. Meek, The effect <strong>of</strong> early calfhood health status on<br />

survivorship and age at first calving, Can. J. Vet. Res. 50 (1986), pp. 314–317.<br />

D. Waltner-Toews, S.W. Martin, A.H. Meek and I. McMillan, Dairy calf management, morbidity<br />

and mortality in Ontario Holstein herds. I. The data, Prev. Vet. Med. 4 (1986), pp. 103–124.<br />

W.D. Yates, A review <strong>of</strong> infectious bovine rhinotracheitis, shipping fever pneumonia and viralbacterial<br />

synergism in respiratory disease <strong>of</strong> cattle, Can. J. Comp. Med. 46 (1982), pp. 225–263.<br />

151


Table 6.1: Disease variables considered in models for weight, height, and risk <strong>of</strong> bovine<br />

respiratory disease (BRD)<br />

Disease event Clinical Signs<br />

BRD Dull and listless with a rectal temperature > 39.5°C, and/or elevated<br />

respiratory rate and/or nasal discharge<br />

Scours Feces are s<strong>of</strong>t and do not hold form<br />

Otitis media One or both ears are drooping or head is tilted and calf is unable to<br />

correct position<br />

Lameness Abnormal gait with unknown cause<br />

Navel ill Navel is swollen or tender in newborn calf<br />

152


Table 6.2: The probability <strong>of</strong> respiratory disease in the 6 wk after movement in 1,392 weaned<br />

dairy calves, randomly assigned to treatment with tulathromycin (TUL) or oxytetracycline (TET)<br />

at the time <strong>of</strong> movement to group housing 1<br />

Variable Estimate SE Odds ratio 95% CI P-value<br />

Intercept −0.44 0.34<br />

Treatment with TUL −0.68 0.15 0.51 0.38–0.68 < 0.001<br />

(relative to TET)<br />

Pre-enrollment lameness<br />

(relative to calves without<br />

lameness)<br />

Weight gain pre-enrollment<br />

(kg/d)<br />

−0.85 0.47 2.35 0.93–5.94 0.07<br />

−1.53 0.57 0.007<br />

1 Logistic regression model, accounting for source farms and weekly enrollment cohort with<br />

random effects.<br />

153


Table 6.3: Least squares means ( ± SEM) <strong>of</strong> the weight (kg) <strong>of</strong> commercial heifers recorded<br />

upon exit from the weaning barn controlling for age at <strong>final</strong> measurement and weight at<br />

enrollment with random effects for source farm and weekly enrollment cohort<br />

Treatment 2<br />

Preweaning BRD 1 n TUL TET P-value<br />

Yes 190 105.5 ± 1.4 104.4 ± 1.7 0.38<br />

No 1,181 110.6 ± 1.1 105.7 ± 1.1 < 0.001<br />

1 Bovine respiratory disease.<br />

2 Treatment with tulathromycin (TUL) or oxytetracycline (TET) at the time <strong>of</strong> movement to<br />

group housing.<br />

154


Table 6.4: Least squares means ( ± SEM) <strong>of</strong> the height (cm) <strong>of</strong> commercial heifers upon exit<br />

from the weaning barn controlling for age at <strong>final</strong> measurement and height at enrollment with<br />

random effects for source farm and weekly enrollment cohort<br />

Treatment 1<br />

Health Event N TUL TET P-value<br />

Preweaning scours<br />

Yes 872 97.8 ± 0.3 97.7 ± 0.3 0.71<br />

No 499 98.3 ± 0.4 97.6 ± 0.4 < 0.005<br />

Preweaning BRD 2<br />

Yes 190 97.5 ± 0.4 97.5 ± 0.4 0.97<br />

No 1,181 98.6 ± 0.3 97.8 ± 0.3 < 0.001<br />

1 Treatment with tulathromycin (TUL) or oxytetracycline (TET) at the time <strong>of</strong> movement to<br />

group housing.<br />

2 Bovine respiratory disease.<br />

155


Table 6.5: Reasons for death loss and culling in the 60 d following movement to group housing<br />

Treatment group 1 BRD 2 confirmed by BRD diagnosed by farm Other Total deaths<br />

digital necropsy staff (unconfirmed)<br />

TET 6 5 5 16<br />

TUL 3 4 4 11<br />

1 Treatment with tulathromycin (TUL) or oxytetracycline (TET) at the time <strong>of</strong> movement to<br />

group housing.<br />

2 Bovine respiratory disease.<br />

156


Figure 6.1: Timeline <strong>of</strong> trial events<br />

157


CHAPTER 7: THE IMPACT OF BOVINE RESPIRATORY DISEASE AND A<br />

PREVENTATIVE ANTIBIOTIC TREATMENT ON GROWTH, SURVIVAL, AGE<br />

AT FIRST CALVING, AND PRODUCTIVITY HEIFERS<br />

7. 1 ABSTRACT<br />

Bovine respiratory disease complex (BRD) is a common disease in weaned dairy calves<br />

that incurs economic and welfare costs. This study was an extension <strong>of</strong> a randomized clinical<br />

trial in which a single injection <strong>of</strong> tulathromycin (TUL) or oxytetracycline (TET) was<br />

administered at first movement to group housing for the prevention <strong>of</strong> BRD in the 60 days<br />

following antimicrobial treatment (BRD60). Calves treated with TUL were 0.5 times (95% CI:<br />

0.4 to 0.7) as likely to be treated for BRD60 as calves treated with TET. The objectives <strong>of</strong> the<br />

current study were to evaluate the long-term effects <strong>of</strong> BRD and antibiotic treatment on growth<br />

<strong>of</strong> heifers until breeding age, age at first calving, incidence <strong>of</strong> dystocia, milk production and<br />

mortality prior to first calving and mortality prior to 120 days in milk. Body weight and average<br />

daily gain were analyzed using generalized linear mixed models (Proc Mixed, SAS) controlling<br />

for source farm and enrolment cohort as random effects. Survival to calving and calving prior to<br />

25 months were analyzed using mixed models with binary distribution and logit link function<br />

(Proc GLIMMIX) in SAS (Version 9.1). A p-value <strong>of</strong> less than 0.05 was considered to be<br />

statistically significant. At entry to the breeding barn (382 days <strong>of</strong> age) calves that experience<br />

BRD60 weighed 16.0 ± 2.3 kg less than calves that did not. For TET and TUL calves, calves<br />

without BRD60 were 3.4 {Confidence Interval: 2.2-5.2} and 1.7 {CI:1.0-2.9} times more likely<br />

to survive to first calving compared to BRD60 calves, respectively. BRD60 calves were 0.6<br />

(95% CI: 0.4-0.8) times as likely to calve before 25 months <strong>of</strong> age and 1.5 (95% CI: 1.1-2.2)<br />

times more likely to have a calving ease score ≥ 2 at their first calving. BRD may affect first test<br />

158


milk production depending on farm management factors, but had no effect on projected 305 day<br />

milk production. BRD has many long-term effects on the productivity <strong>of</strong> heifer calves. The<br />

administration <strong>of</strong> TUL at movement to group housing may have a role in the prevention <strong>of</strong> BRD<br />

and mediating some <strong>of</strong> the long-term effects <strong>of</strong> this disease.<br />

159


7.2 INTRODUCTION<br />

Bovine respiratory disease complex (BRD) is <strong>of</strong> great concern due to the high economic<br />

and welfare costs associated with this condition. According to the 2007 U.S. National Animal<br />

Health Monitoring Survey (NAHMS) the incidence <strong>of</strong> post-weaning respiratory disease in dairy<br />

heifers was 5.9%, and was the predominant cause <strong>of</strong> reported deaths <strong>of</strong> weaned heifers (46.5%)<br />

(USDA, 2010). This disease is <strong>of</strong>ten observed following first movement to group housing after<br />

weaning (McGuirk, 2008). Risk factors for BRD include concurrent nutritional, environmental<br />

and social changes, as well as increased disease exposure from mixing with older animals for the<br />

first time.<br />

Severe, undetected or untreated BRD may result in death. Prompt therapy should result in<br />

a low case fatality risk, but this disease can represent an economic drain for producers. In the<br />

short-term, costs <strong>of</strong> the disease include labor for detection and treatment <strong>of</strong> disease, drugs,<br />

veterinary fees and reduced performance and cost to replace animals that are died as a result <strong>of</strong><br />

clinical disease. Costs <strong>of</strong> this disease were estimated at $14.71(range: 0 - $119) per pre-weaned<br />

calf per year and $1.95 (range: 0-$9.25) per weaned dairy calf stock per year by Michigan<br />

producers in 1990 (Kaneene and Hurd, 1990). The higher cost observed in young calves is based<br />

on the timing <strong>of</strong> preventive practices, which primarily are administered to calves prior to disease<br />

outbreaks, and the long follow-up time for calf stock which included an extended period <strong>of</strong> low<br />

disease risk. However, the long-term costs <strong>of</strong> BRD are likely underestimated, since generally<br />

poor disease records available for heifers make it difficult for producers to associate BRD with<br />

outcomes many months later. Costs <strong>of</strong> long-term effects were estimated between 18-57 euros ($<br />

15-49 U.S.) in a partial budget <strong>of</strong> the cost <strong>of</strong> BRD on a typical Dutch dairy farms, in which 60%<br />

160


<strong>of</strong> dairy heifers were affected by mild to chronic BRD (van der Fels-Klerx et al., 2001).<br />

However, the effects <strong>of</strong> BRD on growth, survival and production were based on a small number<br />

<strong>of</strong> studies examining these factors. These studies have documented different long-term costs<br />

include increased risk <strong>of</strong> mortality prior to calving (Waltner-Toews, et al, 1986), decreased<br />

growth (Virtala et al, 1996), later age at calving (Waltner-Toews, et al, 1986; Correa et al, 1988)<br />

and increased risk <strong>of</strong> dystocia (Warnick et al, 1994). Warnick et al, (1995; 1997) found no<br />

association <strong>of</strong> calfhood BRD with first lactation milk production or survival after first calving.<br />

However, calves with more than 4 treatments for BRD prior to calving were 2 times more likely<br />

to survive first lactation compared to calves with no BRD(Bach, 2011). BRD in beef feedlots is<br />

associated with decreased ADG, marbling, meat quality, and subcutaneous fat cover and<br />

increased mortality (Smith, 1998; Reinhardt et al., 2009; Schneider et al., 2009). There is a need<br />

re-examine the effects <strong>of</strong> BRD on dairy heifer growth, age at first calving, dystocia, mortality<br />

and milk production in first lactation due to the likely changes in management practices, BRD<br />

treatments and animal genetics.<br />

This study is an extension <strong>of</strong> a randomized clinical trial, in which a single injection <strong>of</strong><br />

tulathromycin (TUL) or oxyetetracycline (TET) was administered to dairy heifers at first<br />

movement to group housing for the prevention <strong>of</strong> BRD in the 60 days following antimicrobial<br />

treatment (BRD60) (<strong>Stanton</strong> et al., 2010). Briefly, 1,392 calves, raised in a commercial heifer<br />

raising facility were enrolled at first movement to group housing when they were 8 weeks <strong>of</strong> age.<br />

A total <strong>of</strong> 248 calves were treated for BRD in the 60 days following enrollment. Calves assigned<br />

to the TUL group were 0.5 times (95% CI: 0.4 to 0.7) as likely to be treated for BRD in the 60<br />

161


days following study enrolment than calves assigned to the TET group. That study was restricted<br />

to the first 60 days following movement to group housing.<br />

The objectives <strong>of</strong> the present study were to evaluate the long-term effects <strong>of</strong> BRD and<br />

antimicrobial treatment performance. Specifically, growth <strong>of</strong> heifers until breeding age, age at<br />

first calving, incidence <strong>of</strong> dystocia, milk production and mortality prior to first calving and<br />

mortality prior to 120 days in milk were examined.<br />

7.3 MATERIAL AND METHODS<br />

7.3.1 Heifer Management and Housing<br />

Heifers enrolled in this study (1,392) were reared at a commercial heifer raising facility<br />

in Western New York. They arrived at the facility from 7 separate source farms at an average age<br />

<strong>of</strong> 3 ± 2.1 (mean ± SD) days <strong>of</strong> age. Calves were enrolled following movement from individual<br />

housing in hutches or nursery barns during the milk-fed period to group housing post-weaning.<br />

The average age at enrolment was 56 days <strong>of</strong> age. Calves were randomly assigned to treatment<br />

with either 2cc <strong>of</strong> tulathromycin (TUL) or 5cc <strong>of</strong> oxyetetracycline (TET). Calves were monitored<br />

for disease for 60 days following study enrollment. The definition <strong>of</strong> BRD was dullness,<br />

decreased appetite or listlessness, with at least one <strong>of</strong> the following; rectal temperature > 39.5 ºC,<br />

nasal discharge, or elevated respiratory effort. Bovine respiratory disease was diagnosed in<br />

13.2% (92/697) <strong>of</strong> TUL calves and 22.4% (156/695) <strong>of</strong> TET calves during the 60 days following<br />

first movement to group housing.<br />

All heifers remained at the heifer raising facility until they were confirmed pregnant.<br />

After confirmation <strong>of</strong> pregnancy, heifers were eligible to return to their source farm and did so<br />

162


ased on the preset preferences <strong>of</strong> the original source farm. All heifers were returned to their<br />

original source farms unless they died or were culled prior to calving, (Figure 7.1).<br />

7.3.2 Outcome Measures<br />

Body weight, height and average daily gain<br />

All calves were monitored daily in the nursery, weaning and transition barns by trained<br />

barn staff for the occurrence <strong>of</strong> disease, using consistent case definitions. All disease events and<br />

treatments were recorded in the heifer farm management database (DairyComp 305, Valley<br />

Agricultural S<strong>of</strong>tware Inc., Tulane, CA.) prior to returning to the source farms. When heifers<br />

were returned to their source farms to calve and to enter the milking herd, all disease, production,<br />

calving and mortality information were recorded by the farms’ staff using DairyComp 305.<br />

Weights and heights <strong>of</strong> calves were measured at standard times using an electronic scale<br />

(Tru-test - ID3000, Tru-Test Incorporated, Mineral Wells Texas) as part <strong>of</strong> the standard<br />

operating procedure <strong>of</strong> the commercial heifer rearing facility. Height at the withers (to the<br />

nearest inch) was measured at the same time as weight.<br />

Measurements were taken at entry into the following barns; Nursery Barn (mean ± s.d.<br />

age in days; 3 ± 2), Weaning Barn (56 ± 7), Transition Barn (98 ± 8), Grower Barn 1 (181 ± 13),<br />

Grower Barn 2 (269 ± 31) and Breeding Barn (382 ± 33).<br />

Survival to first calving<br />

Survival was deemed to occur if heifers calved in their original source herd.<br />

163


Age at First Calving<br />

Since age at first calving is not normally distributed, median and quartiles are presented<br />

to describe the age at first calving for the population. The outcome age at first calving was<br />

ultimately divided into two categories, heifers that calved up to and including 25 months <strong>of</strong> age<br />

and those that calved after 25 months <strong>of</strong> age. This is based on the widely used economic target <strong>of</strong><br />

a calving age <strong>of</strong> 24 months with an additional month added to allow for minor deviations from<br />

this goal (Le Cozler et al., 2008).<br />

Dystocia<br />

Calving ease was scored in accordance with the standard 5 point system <strong>of</strong> the National<br />

Dairy Herd Information Association (DHIA) as part <strong>of</strong> the source farms’ standard practices<br />

(Table 7.1). Cutpoints <strong>of</strong> calving ease scores ≥2, ≥3, ≥4, and ≥5 were independently assessed as<br />

potential indicators <strong>of</strong> dystocia since there is no validated cutpoint for dystocia using on this<br />

calving ease scale.<br />

Survival to 120 days in milk<br />

Survival to 120 days in milk was deemed to occur if heifers were present in their source<br />

herd for ≥120 days after their first calving.<br />

Milk Production<br />

Regular DHIA milk testing was performed at monthly intervals on 4 <strong>of</strong> the 6 farms and<br />

every other month on the remaining farms. First test milk production and third test projected<br />

305- day milk production were collected for the 4 farms using monthly testing. First test milk<br />

production and second test projected 305 –day milk production were collected for the 2 farms<br />

164


that were testing every other month in an attempt to standardize days in milk at test days among<br />

the 6 farms.<br />

Culling and Mortality<br />

The causes <strong>of</strong> culling and mortality were entered into DairyComp305 by the staff <strong>of</strong> the<br />

heifer raising facility or <strong>of</strong> the individual source farms, depending on the location <strong>of</strong> the animal<br />

at the time the determination was made. Producer-attributed cause <strong>of</strong> death or culling was<br />

recorded as respiratory disease, injury, other, unknown. Culling was divided into the following<br />

categories based on the primary reason given by producers: respiratory, other diseases, sold for<br />

dairy, reproduction, injury, poor growth, other.<br />

7.3.3 Statistical Analysis<br />

All analyses were conducted using SAS statistical s<strong>of</strong>tware Version 9.1 (SAS Institute<br />

Inc., Cary, NC). For all models, statistical significance was based on a p-value ≤ 0.05.<br />

Season was <strong>of</strong>fered to all models as a fixed effect. Season was divided into three periods<br />

based on a visual assessment <strong>of</strong> daily temperature recordings obtained from the Rochester, New<br />

York Weather Station (Weather Underground, 2010) (Appendix 1): Moderate (5 ± 4 o C (mean ±<br />

standard deviation); November 1 st , 2006 to January 15 th , 2007); Cold (-2 ± 7 o C: January 16 th to<br />

April 15 th , 2007); Warm (18 ± 6 o C; April 16 th , 2007 to the end <strong>of</strong> the trial in July 2007). The<br />

Rochester Weather Station is located approximately 65 km from the heifer raising facility.<br />

Body weight, height and average daily gain<br />

The impacts <strong>of</strong> antimicrobial treatment and BRD60 on body weight and height were<br />

evaluated separately at each management time point, corresponding to entry into Rearing Barns<br />

165


(Weaning, Transition, Grower 1, Grower 2, and Breeding). The impact <strong>of</strong> antimicrobial<br />

treatment and BRD60 on ADG were evaluated separately for each time period corresponding o<br />

time in the Rearing Barns (Weaning, Transition, Grower 1, and Grower 2). In addition, the<br />

impact <strong>of</strong> antimicrobial treatment and BRD60 on ADG between the Weaning Barn and the<br />

Breeding was evaluated. The impact <strong>of</strong> antimicrobial treatment on ADG, weight and height was<br />

evaluated separately from BRD60 since BRD60 is an intervening variable for antimicrobial<br />

treatment and growth.<br />

Due to inconsistent recording <strong>of</strong> breeding barn exit weights, the ADG <strong>of</strong> heifers in the<br />

breeding barn were not analyzed. Body weight, height and ADG were evaluated using linear<br />

regression mixed models (PROC MIXED) with random effects for source farm and weekly<br />

enrolment cohort. The models for body weight controlled for birth weight, pre-enrollment<br />

average daily gain (pADG), pre-enrolment respiratory disease (pBRD), season and age at entry<br />

into the rearing barn. The models <strong>of</strong> height controlled for height at birth, pBRD, and age at entry<br />

into the barn. The models <strong>of</strong> ADG controlled for birth weight, pBRD, pADG, season and age at<br />

exit from barn.<br />

Dystocia, survival to first calving, and first lactation survival (120 days)<br />

First lactation survival to 120 days following calving, as well as dichotomous outcomes<br />

for dystocia at various cutpoints, age category at first calving and survival to first calving, were<br />

analyzed using mixed models with binary distribution and logit link function (Proc GLIMMIX).<br />

Source farm and weekly enrollment cohort were included as random effects. Antimicrobial<br />

treatment, BRD, pBRD and season were included as fixed effects. In addition, the association<br />

between age at first calving and calving ease with first lactation survival was modeled, without<br />

166


BRD or antimicrobial treatment, since these are intervening variables between BRD and<br />

survival.<br />

Age at First Calving<br />

Time to first calving was analyzed using a Cox Proportional Hazard survival analysis<br />

model (PROC tphreg), controlling for source farm and group as random effects, and using a non-<br />

parametric estimate <strong>of</strong> the survival distribution function (PROC Lifetest).<br />

Milk production<br />

Milk production, first test DHIA milk weight and 305d milk, were analyzed using linear<br />

regression mixed models (PROC MIXED) with random effects for source farm and weekly<br />

enrolment cohort. Two models were built for milk production. The first model evaluated the<br />

impact <strong>of</strong> BRD60 and antimicrobial treatment on milk production, and the second model<br />

evaluated age at first calving and calving ease score. These two sets <strong>of</strong> predictors were modeled<br />

separately since age at first calving and calving ease score are intervening variables for BRD60.<br />

The models for milk production controlled for days in milk at DHIA testing and source farm as<br />

fixed effects. Source farm was included as a fixed effect in this model to determine if<br />

management factors were interacting with milk production. Residual diagnostics and normality<br />

was assessed using Proc Univariate.<br />

7.3.4 Exclusion Criteria<br />

One source farm did not provide records after return to the farm for calving. Animals<br />

from this farm were excluded from analysis <strong>of</strong> survival to first calving and from analyses <strong>of</strong> all<br />

events that occurred after calving. These calves were included in analysis <strong>of</strong> average daily gain.<br />

167


The excluded source farm resulted in a loss <strong>of</strong> 3% <strong>of</strong> calves from the original database<br />

(48/1393).<br />

7.4 RESULTS<br />

The weights and heights <strong>of</strong> calves upon entry to the six rearing barns are presented in<br />

Table 7.2. Four heifers were not weighed or measured at one <strong>of</strong> the measurement time points<br />

(transition barn (n = 1), grower barn 1 (n = 1) and grower barn 2 (n = 2)) and were excluded<br />

from all analyses <strong>of</strong> ADG, weight and height that require the use <strong>of</strong> this data.<br />

7.4.1 Average Daily Gain<br />

Effects <strong>of</strong> BRD60 on ADG are presented in Table 7.3. BRD60 was associated with a 0.16<br />

± 0.01 kg decrease in ADG in the weaning barn (P < 0.001). Similarly, BRD60 was associated<br />

with reduced ADG in the Transition Barn and Grower Barn 1. Overall, BRD60 was associated<br />

with a decrease in ADG between the weaning barn and the breeding <strong>of</strong> 0.03 ± 0.01 kg (P <<br />

0.001).<br />

An antimicrobial treatment and pBRD interaction was significant in the weaning and<br />

transition barns, (P < 0.005 and P=0.05, respectively). The interaction between antimicrobial<br />

treatment and pBRD did not have a significant effect on ADG in Grower Barn 1 or Grower Barn<br />

2, (P = 0.51 and P = 0.25, respectively). In the weaning barn, calves without pBRD gained 0.97<br />

± 0.01 kg/day and 0.86 ± 0.01 kg/day in the TUL and TET group, respectively (P < 0.001). In<br />

the transition barn, calves without pBRD gained 0.98 ± 0.01 kg/day and 0.96 ± 0.01 kg/day in<br />

the TUL and TET group, respectively, (P = 0.03). In calves with pBRD, antimicrobial treatment<br />

had no effect on ADG, (P = 0.57).<br />

168


7.4.2 Body weight<br />

Calves treated for BRD60 had lower body weights starting at entry into the transition<br />

barn (3.5 months) and continuing until the last measurement at entry into the breeding barn, (P <<br />

0.001; Figure 7.2). The maximum difference in body weight observed between BRD60 and non-<br />

BRD60 calves was 15.4 ± 2.3 kg upon entry to the grower barn 2 at 269 days <strong>of</strong> age (P < 0.001).<br />

Antimicrobial treatment and pBRD interactions were significant for all weights taken<br />

after entry into the transition barn, (P < 0.05). For calves with no pBRD, the greatest difference<br />

in body weight observed between TUL and TET calves was 7.3 ± 1.4 kg, which occurred at entry<br />

into Grower Barn 2 at 269 days <strong>of</strong> age (P < 0.001; Figure 7.3).<br />

7.4.3 Height<br />

Calves treated for BRD60 were 1.3 ± 0.2, 2.0 ± 0.2, 1.5 ± 0.2, and 1.6 ± 0.4 cm smaller<br />

than calves without BRD60 when measured at entry into the Transition Barn, Grower Barn 1,<br />

Grower Barn 2 and Breeding Barn, respectively (P < 0.001).<br />

In grower barn 1 and breeding barn, antimicrobial treatment and pBRD interacted (P =<br />

0.01 and P = 0.0003, respectively). For calves without pBRD, TUL calves were 1.0 ± 0.2 cm<br />

taller than TET calves in grower barn 1, (P < 0.001; Table 7.4). In grower barn 2, pBRD and<br />

antimicrobial treatment were not significant (P = 0.61) and TUL calves were 1.1 ± 0.3 cm taller<br />

than TET calves (P < 0.001).<br />

7.4.4 Survival to First Calving<br />

One heifer returned to her source farm but was lost to follow-up prior to calving. A total<br />

<strong>of</strong> 1,088/1343 (81%) heifers survived to first calving at their source farms. Only 66% (158/238)<br />

169


<strong>of</strong> calves with BRD60 survived to first calving and 84% (931/1105) <strong>of</strong> calves without BRD60<br />

survived to first calving. Survival to first calving was 83% (559/676) and 79% (530/667) for<br />

TUL and TET calves, respectively. Of the 237 calves with BRD60, 73% (64/88) <strong>of</strong> the TUL<br />

calves survived to first calving and 62% (93/149) <strong>of</strong> the TET calves survived to first calving.<br />

Survival to first calving was affected by BRD60 and antimicrobial treatment with a<br />

significant interaction between antimicrobial treatment and BRD60, P < 0.03. For TET and TUL<br />

calves, calves without BRD60 were 3.4 (2.2-5.2) and 1.7 (1.0-2.9) times more likely to survive to<br />

first calving compared to BRD60 calves, respectively. Calves treated for BRD60 were 2.0 (95%<br />

CI: 1.1-3.6) times more likely to survive to first calving if they had received TUL at movement<br />

to group housing compared to TET. Antimicrobial treatment did not have an effect on the<br />

probability <strong>of</strong> non-BRD60 calves surviving to first calving, (OR: 1.0; CI: 0.7-1.5, P = 0.80).<br />

Survival to first calving by source farm is reported in Table 7.5. There was no interaction <strong>of</strong> farm<br />

with the effect <strong>of</strong> BRD60 on survival.<br />

7.4.5 Culling and mortality prior to first calving<br />

Of the 255 heifers that did not survive to first calving, 36% (93/255) died and 64%<br />

(162/255) were culled. The primary reason for mortality was BRD, with 47% (44/93) <strong>of</strong> all<br />

deaths attributed this disease (Table 7.6). For the deaths attributed to BRD, 61% (27/44) were<br />

treated for BRD in the 60 days following movement to group housing with 34% (15/44) and 66%<br />

(29/44) <strong>of</strong> deaths due to BRD in the TUL and TET groups, respectively (P = 0.04). Poor<br />

reproduction was the primary reason for culling <strong>of</strong> heifers (Table 7.7).<br />

170


7.4.6 Age at First Calving<br />

Of the 1,088 heifers that returned to calve at their source farms, the median age at first<br />

calving was 703 (First and third quartile; 683-737) days. The median age at first calving was 714<br />

(q1 - q3; 705-723) and 702 (699 - 706) days for BRD60 and non-BRD60 calves, respectively.<br />

Only 19% (205/1088) <strong>of</strong> heifers failed to calve prior to 25 months <strong>of</strong> age. The proportion <strong>of</strong><br />

BRD60 and non-BRD60 heifers that failed to calve prior to 25 months <strong>of</strong> age was 27% (43/158)<br />

and 17% (162/931), respectively. Controlling for source farm, enrollment cohort and<br />

antimicrobial treatment, calves with BRD60 were 0.6 (0.4-0.8) times as likely to calve by 25<br />

months <strong>of</strong> age (P = 0.01). There was no association (P = 0.98) <strong>of</strong> antimicrobial treatment with<br />

probability <strong>of</strong> calving by 25 months <strong>of</strong> age. Based on the results <strong>of</strong> the survival analysis, heifers<br />

with BRD60 were 0.8 (95% CI: 0.7-1.0) times as likely to calve as heifers without BRD60 (P =<br />

0.02; Figure 7.4).<br />

7.4.7 Dystocia<br />

Calving ease scores were recorded for 98% (1066/1089) <strong>of</strong> heifers which calved at their<br />

source farms (Table 7.8). A calving ease score ≥2 was recorded for 48% (509/1066) <strong>of</strong> heifers.<br />

Fifty-five percent <strong>of</strong> the heifers with BRD60 had a calving ease score ≥2 (85/155) compared to<br />

46% (424/911) <strong>of</strong> calves without BRD60. Controlling for source farm, enrollment cohort and<br />

antimicrobial treatment, calves with BRD60 were 1.5 (95%CI: 1.1-2.2) times more likely to have<br />

a calving ease score ≥ 2 at their first calving than calves without BRD60 (P = 0.03). However,<br />

BRD60 was not associated with the probability <strong>of</strong> calving scores ≥ 3 (Table 7.9). There was no<br />

association <strong>of</strong> antimicrobial treatment with the probability <strong>of</strong> a calving score ≥ 2, ≥3, or ≥4 (P =<br />

0.71, P = 0.98, and P = 0.18, respectively).<br />

171


7.4.8 Milk Production<br />

First Test Day Milk production<br />

A total <strong>of</strong> 1,041 heifers had first DHIA test milk production. Eleven heifers were<br />

removed from this analysis due to abnormalities identified by the producer at the time <strong>of</strong> milk<br />

testing. There was a significant interaction <strong>of</strong> source farm with the effect <strong>of</strong> BRD60 on first test<br />

milk production (P = 0.04). For two <strong>of</strong> the farms, heifers with BRD60 had significantly lower<br />

milk production than heifers without BRD60 (Table 7.10). For one <strong>of</strong> the source farms, heifers<br />

with BRD60 had significantly higher milk production than heifers without BRD60 (P = 0.05).<br />

For the remaining three source farms there was no statistically significant difference in milk<br />

production.<br />

Calving ease and age at first calving were both associated with milk production at first<br />

milk test, (P = 0.1 and 0.7, respectively). Average first test milk production for calving ease<br />

scores 1,2,3,4 and 5 were 28.8 ± 0.3, 28.8 ± 0.4, 27.8 ± 0.6, 27.6 ± 0.9 and 26.0 ± 1.2 kg,<br />

respectively. Heifers with a calving ease score <strong>of</strong> 1 and 2 had significantly higher milk<br />

production at first test compared to heifers with a calving ease score <strong>of</strong> 5 (P < 0.05).<br />

Projected 305 D milk production<br />

A total <strong>of</strong> 1,019 cows had projected 305d milk. For every day age at first calving<br />

increased, 305d milk production decreased by 5.2 ± 1.3 (lsmeans ± S.E.M.) kg. BRD60 and<br />

antimicrobial treatment were not significantly associated with 305d milk production (P = 0.75<br />

and P = 0.46, respectively). Calving ease score was not significantly associated with 305d milk<br />

production (P = 0.92).<br />

172


7.4.9 Survival to 120 days in milk<br />

Twenty-four animals (2% <strong>of</strong> total) were lost to follow-up or did not complete 120 days in<br />

milk. Of the animals that survived to first calving, 92% (1001/1064) survived to 120 days in<br />

milk. BRD60 and antimicrobial treatment did not significantly affect the survival to 120 days in<br />

milk, (P = 0.94 and P = 0.36, respectively). Age at first calving was not associated with post-<br />

calving survival (P = 0.70). Calving ease scores were recorded for 1,042 <strong>of</strong> the heifers that<br />

survived to their first calving. Calving ease score was associated with post-calving survival (P =<br />

0.04). Heifers with a calving ease score <strong>of</strong> 5 were 0.2 (95% CI: 0.1-0.9) times as likely to survive<br />

to 120 days post-calving compared to heifers with a calving ease score <strong>of</strong> 1 (P = 0.04; Table<br />

7.11).<br />

7.5 DISCUSSION<br />

This study evaluated the long-term effects <strong>of</strong> BRD under modern management conditions<br />

and the long-term effects <strong>of</strong> a novel approach to reducing the incidence and effect <strong>of</strong> BRD in<br />

dairy calves. Specifically, the effect <strong>of</strong> a long-acting prophylactic treatment was compared to a<br />

shorter acting prophylactic treatment. Respiratory disease has previously been documented to<br />

have many long-term effects on the health and productivity <strong>of</strong> dairy heifers. However, these<br />

studies have become dated and may not reflect current management practices.<br />

The negative effects <strong>of</strong> BRD60 following movement to group housing on ADG were<br />

seen until approximately 9 months <strong>of</strong> age, or 7 months after movement to group housing. This<br />

difference in average daily gain resulted in a 15 ± 2 kg decrease in body weight for calves with<br />

BRD60 at entry into the breeding barn at this facility (approximately 382 days <strong>of</strong> age). This<br />

finding is consistent with previous studies which found that that the number <strong>of</strong> days treated for<br />

173


BRD before 6 months <strong>of</strong> age decreased ADG between 6 and 14 months by 0.02 kg/day(Donovan<br />

et al., 1998). However, direct comparison is not possible given the variation in follow-up time<br />

and the differences in outcomes. Donovan et al (1998) looked at number <strong>of</strong> days treated while<br />

this study assessed the presence or absence <strong>of</strong> a BRD event within the first 60 days <strong>of</strong> movement<br />

to group housing. Both <strong>of</strong> these studies indicate that BRD has a negative impact on the growth <strong>of</strong><br />

dairy heifers for many months after initial treatment. This is inconsistent with findings from the<br />

beef feedlot industry that BRD does not have a significant long-term effect on ADG following an<br />

initial clinical phase (Thompson et al., 2006; Holland et al., 2010). This discrepancy may be due<br />

to differences in the stage <strong>of</strong> the animal’s physical development at onset <strong>of</strong> disease, or a<br />

difference in the type or incidence <strong>of</strong> pathogens. In the study by Holland et al, (2010), beef<br />

animals entered the study at 241 kg and they contracted respiratory disease in the pre-<br />

conditioning phase which occurred over the next 63 days. Following the pre-conditioning phase,<br />

there was no effect <strong>of</strong> BRD treatment on ADG. The disparity in size between the calves on the<br />

present study (enrolled at 71kg) may indicate that that the age and weight <strong>of</strong> the animal at<br />

disease onset plays a role on the impact <strong>of</strong> BRD. If the majority <strong>of</strong> structural growth has already<br />

occurred at the onset <strong>of</strong> disease then the long-term impact <strong>of</strong> BRD may be reduced.<br />

BRD60 decreased the height <strong>of</strong> heifers up to entry into the breeding barn by 1.6 ± 0.4 cm<br />

at approximately 382 days <strong>of</strong> age. The maximum observed difference in height between heifers<br />

with and without a recorded case <strong>of</strong> BRD60 was -2.0 ± 0.2 cm at approximately 180 days <strong>of</strong> age,<br />

upon entry into Grower Barn 1. This is consistent with the findings <strong>of</strong> Donovan et al (1998) that<br />

the impact <strong>of</strong> BRD on height gain is primarily seen in the first 6 months <strong>of</strong> life. However, as<br />

shown in this study, the effect on delayed growth in stature can remain beyond 1 year <strong>of</strong> life.<br />

174


For calves without BRD prior to enrolment, calves treated with TUL at first movement to<br />

group housing had a 0.11 ± 0.01 and 0.02 ± 0.01 kg increase in ADG in the weaning and<br />

transition barns relative to calves treated with TET. This difference in ADG was not observed<br />

after calves left the transition barn. This suggests that TUL had a short-term effect on the ADG<br />

<strong>of</strong> dairy calves in calves with no history <strong>of</strong> BRD prior to antimicrobial treatment. This<br />

improvement in body mass was maintained until all surviving heifers entered the Breeding Barn<br />

at 382 days <strong>of</strong> age. For heifers with a history <strong>of</strong> BRD prior to enrolment, antimicrobial treatment<br />

had no effect on growth. This absence <strong>of</strong> antimicrobial treatment effect in calves with a prior<br />

history <strong>of</strong> BRD indicates that antimicrobial treatment does not compensate for an earlier disease<br />

occurrence, likely due to the effect BRD has already had on heifer health and performance. This<br />

indicates that treatment with TUL is most beneficial for animals that are at low risk <strong>of</strong> BRD pre-<br />

movement to group housing and high risk <strong>of</strong> BRD in the period following movement to group<br />

housing.<br />

It should be noted that the comparison between antimicrobial treatments is limited by the<br />

discrepancies in dosages. The dosage <strong>of</strong> oxytetracycline was based on the previously established<br />

farm protocol <strong>of</strong> the commercial heifer raising facility. Calculation <strong>of</strong> the dosage <strong>of</strong><br />

oxytetracycline that would be appropriate based on the weights <strong>of</strong> calves in this study indicates<br />

that the dose <strong>of</strong> TET was lower than is considered to be optimal. As such, the under dosing <strong>of</strong><br />

oxytetracycline may not represent the effects that would be expected with a full dose (<strong>Stanton</strong> et<br />

al., 2010). However, the value <strong>of</strong> the reduced incidence <strong>of</strong> BRD and improved growth show a<br />

significant benefit <strong>of</strong> tulathromycin for improved prevention <strong>of</strong> clinical BRD and some <strong>of</strong> the<br />

long-term consequences <strong>of</strong> BRD.<br />

175


BRD following movement to group housing had a negative effect on the survival <strong>of</strong><br />

heifers to first calving. A reduction in survival to first calving has two main impacts on the<br />

productivity <strong>of</strong> a farm. It reduces the ability <strong>of</strong> producers to make genetic gains by reducing the<br />

pool <strong>of</strong> animals from which to select and retain replacements. Furthermore, for heifers that die<br />

there is a loss <strong>of</strong> the future value <strong>of</strong> the cow, as well as the wasted costs associated with rearing<br />

the animal up to its death without any financial return. As shown in this study, heifers with<br />

BRD60 were 0.3 to 0.6 times as likely to survive to first calving. This is consistent with the<br />

findings <strong>of</strong> Waltner-Toews, et al (1986) which found that heifers treated for BRD in the first 90<br />

days <strong>of</strong> life were 2.5 times more likely to die between 90 days <strong>of</strong> age and calving. These two<br />

studies differ from Curtis et al,(1988) who found that respiratory disease was not associated with<br />

survival to calving. These differences may be due to the severity <strong>of</strong> the disease events<br />

considered. Curtis et al,(1988) included any BRD event regardless <strong>of</strong> treatment, whereas both<br />

this study and Waltner-Toews et al, (1986) only included calves that were ill enough to be<br />

treated.<br />

Calves treated for BRD60 were 2.0 (95% CI: 1.1-3.6) times more likely to survive to first<br />

calving if they had received TUL at movement to group housing compared to TET (i.e. prior to<br />

disease occurrence). A possible explanation for this interaction is that the longer duration <strong>of</strong><br />

activity <strong>of</strong> tulathromycin relative to oxytetracycline may have reduced the severity <strong>of</strong> the<br />

infection <strong>of</strong> BRD60 in some calves. However, the comparison between these two groups must be<br />

taken cautiously due to the lack <strong>of</strong> a true positive control in this trial. However, these results do<br />

suggest that further study is warranted on the effect <strong>of</strong> TUL administered around weaning on the<br />

survival <strong>of</strong> calves that developed BRD.<br />

176


Age at first calving is an important factor for producers since it has a significant impact<br />

on the lifetime productivity <strong>of</strong> heifers (Le Cozler et al., 2008). In 1985, BRD was shown to<br />

increase the average age at first calving by Warnick et al (1994), who found that heifers with<br />

BRD in the first 3 months <strong>of</strong> life calved three months later than heifers without BRD. The<br />

present study found that heifers treated for BRD in the 60 days following movement to group<br />

housing were almost half as likely to calve prior to 25 months ( OR: 0.6; CI:0.4- 0.8) than heifers<br />

without BRD. Waltner-Toews et al (1986) found no effect <strong>of</strong> BRD on age at first calving which<br />

may be due to in part from dichotomizing age at first calving using a cutpoint <strong>of</strong> 30 months, and<br />

a sample size which is approximately half that <strong>of</strong> the present study. In addition, it is likely that<br />

genetic potential <strong>of</strong> dairy animals and management practices in the dairy industry have changed<br />

since the 1980’s.<br />

Dystocia is an important issue in the dairy industry. According to the 2007 NAHMS<br />

report, difficulty in calving occurred in 18% <strong>of</strong> heifers (USDA, 2009). Dystocia is most common<br />

in heifers, and is commonly associated with body weight at calving, age at calving and sex <strong>of</strong> the<br />

calf (Mee, 2008). In addition, calves born from dystocia events are more likely to be stillborn.<br />

Difficulty in calving can cause damage to the reproductive tract <strong>of</strong> the cow, prolong calving and<br />

can result in surgical intervention. Dystocia can increase the risk <strong>of</strong> retained placenta and<br />

displaced abomasum (Correa et al., 1993; Laven and Peters, 1996). Defining dystocia is<br />

challenging and subject to interpretation. In this study, various cutpoints were used on a<br />

standardized scale <strong>of</strong> calving ease scores to assess the impact <strong>of</strong> BRD on dystocia. BRD did not<br />

significantly affect dystocia rates if the calving ease cutpoints were ≥3. However, the small<br />

number <strong>of</strong> animals with calving ease scores ≥ 3 does present an issue with power. In this study<br />

177


calves with BRD60 were 1.5 (95% CI: 1.1-2.2) times more likely to have at least mild dystocia<br />

(calving ease score ≥ 2). This is lower than in Warnick et al, (1994) who found that calves with<br />

BRD were 2.4 times more likely to have dystocia, but does add support to the association <strong>of</strong><br />

calfhood respiratory disease with dystocia.<br />

BRD60 decreased first test milk weight by 2.3 to 2.9 kg for 2 <strong>of</strong> the 6 farms. For 3 <strong>of</strong> the<br />

farms, the effect <strong>of</strong> BRD60 was not significant and on the remaining farm heifers with BRD60<br />

had higher milk production (P = 0.05). This indicates that the effect <strong>of</strong> BRD60 in calves on early<br />

lactation milk production nearly 2 years later depends on management factors at the herd level.<br />

One potential reason for differences by farm is the rate at which BRD positive heifers were<br />

culled prior to entry into the milking herd. It is worth noting that the herd with a tendency for<br />

BRD60 to have a positive effect on milk production had only 54% <strong>of</strong> BRD60 calves surviving to<br />

first calving and 85% <strong>of</strong> calves without BRD60 surviving to first calving. This different rate in<br />

survival between BRD60 and non-BRD60 heifers on this farm increases the risk that only the<br />

superior and well performing animals are retained in this herd, impacting the ability to determine<br />

the association between BRD60 and milk production. This interaction <strong>of</strong> farm with BRD60 was<br />

not significant for the projected 305 day milk production. These results from 305 day milk<br />

production are consistent with a previous study which found no long term effects <strong>of</strong> BRD<br />

following entry into milk production (Warnick et al., 1995). This may be due to many factors<br />

including culling <strong>of</strong> low producers, a high level <strong>of</strong> variation in the estimate <strong>of</strong> 305d milk<br />

production and the large number <strong>of</strong> factors which influence longer term milk production.<br />

Among heifers that calved, BRD60 did not affect survival to 120 days in milk. This could<br />

be due to many factors including selection for healthy animals, such that by the time animals<br />

178


have calved the most severely affected animals had been removed from the herd. Further culling<br />

may occur after 120 days and bears further examination. Alternatively, dystocia significantly<br />

decreased survival to 120 days <strong>of</strong> milk. Therefore, BRD may indirectly increase the risk <strong>of</strong> a<br />

heifer being removed from the herd prior to 120 days in milk through its association with<br />

dystocia.<br />

In conclusion, TUL results in improved growth post-enrolment for calves with no history<br />

<strong>of</strong> BRD, likely through the mechanism <strong>of</strong> prevention <strong>of</strong> subclinical disease. In addition, TUL has<br />

a modifying effect on the impact <strong>of</strong> BRD60 on survival to first calving which may indicate a<br />

reduction in the severity <strong>of</strong> clinical disease. BRD60 had several negative effects on the long-term<br />

productivity and welfare <strong>of</strong> dairy heifers including decreased growth, decreased survival to first<br />

calving, increased age at first calving, increased risk <strong>of</strong> dystocia, and depending on farm<br />

management perhaps a short-term impact on milk production.<br />

7.6 ACKNOWLEDGEMENTS<br />

The authors acknowledge Dr. Suzanne T. Millman and Tina M. Widowski, the staff and<br />

management <strong>of</strong> CY Heifer Farms ( Elba, NY) and participating source farms for their invaluable<br />

contribution to his research. Funding was provided by Pfizer Animal Health (New York, New<br />

York) and the National Science and Engineering Research Council <strong>of</strong> Canada (Ottawa, Ontario)<br />

179


7.7 REFERENCES<br />

Correa, M. T., C. R. Curtis, H. N. Erb and M. E. White. 1988. Effect <strong>of</strong> calfhood morbidity on<br />

age at first calving in New York herds. Prev. Vet. Med. 6:253-262.<br />

Correa, M. T., H. Erb, & J. Scarlett. 1993. Path analysis for seven postpartum disorders <strong>of</strong><br />

Holstein cows. J. Dairy Sci. 76(5): 1305-1312. (Abstr.).<br />

Curtis, C. R., J. M. Scarlett, H. N. Erb and M. E. White. 1988. Path model <strong>of</strong> individual-calf risk<br />

factors for calfhood morbidity and mortality in New York Holstein herds. Prev Vet Med. 6:43-<br />

62.<br />

Donovan, G. A., I. R. Dohoo, D. M. Montgomery and F. L. Bennett. 1998. Calf and disease<br />

factors affecting growth in female Holstein calves in Florida, USA. Preventive Veterinary<br />

Medicine. 33:1-10.<br />

Holland, B. P., L. O. Burciaga-Robles, D. L. VanOverbeke, J. N. Shook, D. L. Step, C. J.<br />

Richards and C. R. Krehbiel. 2010. Effect <strong>of</strong> bovine respiratory disease during preconditioning<br />

on subsequent feedlot performance, carcass characteristics, and beef attributes. J. Anim. Sci.<br />

88:2486-2499.<br />

Kaneene, J. B. and H. Scott Hurd. 1990. The national animal health monitoring system in<br />

Michigan. III. cost estimates <strong>of</strong> selected dairy cattle diseases. Preventive Veterinary Medicine.<br />

8:127-140.<br />

Laven, R. A. and A. R. Peters. 1996. Bovine retained placenta: Aetiology, pathogenesis and<br />

economic loss. Vet. Rec. 139:465-471.<br />

Le Cozler, Y., V. Lollivier, P. Lacasse and C. Disenhaus. 2008. Rearing strategy and optimizing<br />

first-calving targets in dairy heifers: A review. Animal. 2:1393.<br />

McGuirk, S. M. 2008. Disease management <strong>of</strong> dairy calves and heifers. Vet. Clin. Food. Anim.<br />

24:139-153.<br />

Mee, J. F. 2008. Prevalence and risk factors for dystocia in dairy cattle: A review. The<br />

Veterinary Journal. 176:93-101.<br />

Reinhardt, C. D., W. D. Busby and L. R. Corah. 2009. Relationship <strong>of</strong> various incoming cattle<br />

traits with feedlot performance and carcass traits. J. Anim. Sci. 87:3030-3042.<br />

Schneider, M. J., R. G. Tait Jr., W. D. Busby and J. M. Reecy. 2009. An evaluation <strong>of</strong> bovine<br />

respiratory disease complex in feedlot cattle: Impact on performance and carcass traits using<br />

treatment records and lung lesion scores. J. Anim. Sci. ; J. Anim. Sci. 87:1821-1827.<br />

180


Smith, R. A. 1998. Impact <strong>of</strong> disease on feedlot performance: A review. J. Anim. Sci. 76:272-<br />

274.<br />

<strong>Stanton</strong>, A. L., D. F. Kelton, S. J. LeBlanc, S. T. Millman, J. Wormuth, R. T. Dingwell and K. E.<br />

Leslie. 2010. The effect <strong>of</strong> treatment with long-acting antibiotic at postweaning movement on<br />

respiratory disease and on growth in commercial dairy calves. J. Dairy Sci. 93:574-581.<br />

Thompson, P. N., A. Stone and W. A. Schultheiss. 2006. Use <strong>of</strong> treatment records and lung<br />

lesion scoring to estimate the effect <strong>of</strong> respiratory disease on growth during early and late<br />

finishing periods in South African feedlot cattle. J. Anim. Sci. 84:488-498.<br />

USDA. 2010. Dairy 2007, heifer calf health and management practices on U.S. dairy operations,<br />

2007.#550.0110:.<br />

USDA. 2009. Dairy 2007, part IV: Reference <strong>of</strong> dairy cattle health and management practices in<br />

the united states, 2007.#N494.0209:.<br />

van der Fels-Klerx, H. J., J. T. Sorensen, A. W. Jalvingh and R. B. Huirne. 2001. An economic<br />

model to calculate farm-specific losses due to bovine respiratory disease in dairy heifers. Prev.<br />

Vet. Med. 51:75-94.<br />

Virtala, A. M. K., G. D. Mechor and Y. T. Grohn. 1996. The effect <strong>of</strong> calfhood diseases on<br />

growth <strong>of</strong> female dairy calves during the first 3 months <strong>of</strong> life in New York State. Journal <strong>of</strong><br />

Dairy Science. 79:1040-1049.<br />

Waltner-Toews, D., S. W. Martin and A. H. Meek. 1986. The effect <strong>of</strong> early calfhood health<br />

status on survivorship and age at first calving. Can. J. Vet. Res. 50:314-317.<br />

Warnick, L. D., H. N. Erb and M. E. White. 1994. The association <strong>of</strong> calfhood morbidity with<br />

first-lactation calving age and dystocia in New York Holstein herds. Kenya Vet. 18:177-179.<br />

Warnick, L. D., H. N. Erb and M. E. White. 1997. The relationship <strong>of</strong> calfhood morbidity with<br />

survival after calving in 25 New York Holstein herds. Prev. Vet. Med. 31:263-273.<br />

Warnick, L. D., H. N. Erb and M. E. White. 1995. Lack <strong>of</strong> association between calf morbidity<br />

and subsequent first lactation milk production in 25 New York Holstein herds. J. Dairy Sci.<br />

78:2819-2830.<br />

Underground Weather.2010. Historical records from Rochester New York Weather Station.<br />

(www.wunderground.com)<br />

181


Table 7.1: Calving Ease Scoring System recorded in farm DairyComp records and reported to<br />

DHIA<br />

Score Definition<br />

1 No problem<br />

2 Slight Problem<br />

3 Needed Assistance<br />

4 Considerable Force<br />

5 Extreme Difficulty<br />

182


Table 7.2: Weight and height <strong>of</strong> 1,392 calves upon entry in the 6 rearing barns at the commercial<br />

heifer raising facility (mean ± S.D.)<br />

Rearing Barn N Age (days) Weight (kg) Height (cm)<br />

Nursery Barn (arrival) 1,392 3 ± 2 40 ± 5 80 ± 3<br />

Weaning Barn 1,392 56 ± 7 71 ± 11 90 ± 4<br />

Transition Barn 1,373 98 ± 9 109 ± 16 98 ± 4<br />

Grower Barn 1 1,333 181 ±13 190 ± 28 109 ± 4<br />

Grower Barn 2 1,301 270 ± 34 273 ± 41 117 ± 4<br />

Breeding Barn 1,271 382 ± 33 370 ± 39 126 ± 6<br />

183


Table 7.3: Least-square means (± SE) <strong>of</strong> the average daily gain (kg/day) <strong>of</strong> 1,392 heifers<br />

recorded during housing in 4 barns and between enrolment and exit from Grower Barn 2,<br />

controlling for age at exit from each barn, birth weight, pre-enrolment average daily gain, preenrolment<br />

bovine respiratory disease and season with random effects for source farm and weekly<br />

enrolment cohort.<br />

Average Daily Gain (kg/day)<br />

Growing Period BRD60- a<br />

BRD60+<br />

(n)<br />

b<br />

Difference <strong>of</strong> least -<br />

square mean<br />

P-value<br />

(n)<br />

(BRD60-<br />

BRD60+)<br />

Weaning Barn 0.91 ± 0.01 0.73 ± 0.02 0.16 ± 0.01 < 0.001<br />

(1,133) (240)<br />

Transition Barn 0.97 ± 0.01 0.90 ± 0.02 0.06 ± 0.01 < 0.001<br />

(1,119) (213)<br />

Grower Barn 1 0.93 ± 0.01 0.89 ± 0.02 0.04 ± 0.01 < 0.005<br />

(1,093) (207)<br />

Grower Barn 2 0.87 ± 0.01 0.87 ± 0.02 > 0.01 ± 0.01 0.80<br />

(1,071) (198)<br />

Between enrolment and 0.88 ± 0.01 0.84 ± 0.01 0.04 ± 0.01 < 0.001<br />

exit from Grower Barn 2 (1,072) (199)<br />

a BRD60-: No clinical bovine respiratory disease complex in the 60 days following enrolment<br />

b BRD60+: treated for clinical bovine respiratory disease complex in the 60 days following<br />

enrolment<br />

184


Table 7.4: Differences <strong>of</strong> least squares means (± SE) in the wither height (cm) between calves<br />

administered tulathromycin or oxytetracycline at enrolment at a commercial heifer raising<br />

facility. The linear mixed model controlled for age at <strong>final</strong> measurement, bovine respiratory<br />

disease prior to trial enrolment (pBRD), height at birth, season, and random effects for source<br />

farm and weekly enrollment cohort<br />

Rearing Barn P-value<br />

for interaction term<br />

185<br />

Difference between<br />

tulathromycin and<br />

oxytetracycline groups<br />

P-value<br />

Transition<br />

Barn<br />

0.08<br />

No pBRD 0.7 ± 0.2<br />

pBRD -0.1 ± 0.4<br />

< 0.001<br />

0.80<br />

Grower Barn 1<br />

0.01<br />

No pBRD 1.0 ± 0.2 < 0.0001<br />

pBRD 0.3 ± 0.5<br />

0.60<br />

Grower Barn 2 0.61 1.1 ± 0.3 < 0.001<br />

No pBRD 0.5 ± 0.3 0.16<br />

Breeding Barn < 0.005<br />

pBRD -3.0± 0.9<br />

< 0.001


Table 7.5: Proportion <strong>of</strong> calves surviving to first calving by source farm and bovine respiratory<br />

disease status<br />

Source Farm BRD60 + 1 BRD60- 2<br />

1 64% (37/58) 80% (238/297)<br />

2 87% (20/23) 91% (77/85)<br />

3 61% (19/31) 92% (132/143)<br />

4 57% (13/23) 67% (87/129)<br />

5 71% (55/77) 89% (307/346)<br />

6 54% (14/26) 85% (89/105)<br />

1 Calves treated for bovine respiratory disease in the 60 days following enrolment<br />

2 Calves without bovine respiratory disease in the 60 days following enrolment<br />

186


Table 7.6: Producers’ primary reason for the 7% (93/1293) <strong>of</strong> heifers that died between<br />

enrolment (8weeks <strong>of</strong> age) and calving.<br />

Producer Attributed Cause <strong>of</strong> Death Proportion <strong>of</strong> Total Deaths<br />

Respiratory Disease 47% (44 /93)<br />

Injury 27% (25/93)<br />

Non-respiratory Diseases 14% (13/93)<br />

Other 1% (1/93)<br />

Unknown 11% (10/93)<br />

187


Table 7.7: Producers primary reason for the 12% (162/1,392) <strong>of</strong> heifers culled from herd<br />

between enrolment (8weeks <strong>of</strong> age) and calving.<br />

Primary Reason for Culling Proportion <strong>of</strong> Cullings<br />

Poor reproduction 32% (51/162)<br />

Poor growth 22% (36/162)<br />

Dairy 22% (35/162)<br />

Non-respiratory diseases 16% (26/162)<br />

Injury 4% (6/162)<br />

Bovine Respiratory Disease Complex 2% (3/162)<br />

Other or unknown 3% (5/162)<br />

188


Table 7.8: Distribution <strong>of</strong> calving ease score recorded by source farms in 1066 heifers<br />

Score Definition Proportion<br />

1 No problem 52% (557)<br />

2 Slight Problem 24% (259)<br />

3 Needed Assistance 14% (151)<br />

4 Considerable Force 6% (60)<br />

5 Extreme Difficulty 4% (39)<br />

189


Table 7.9: Effect <strong>of</strong> BRD60 on calving ease score using multiple cut-points as evaluated in a<br />

logistic model with random effects for source farm and enrolment cohort.<br />

Dystocia Incidence <strong>of</strong> Dystocia Events Odds Ratio : 95% P-value<br />

BRD + BRD-<br />

Confidence Interval<br />

cut-points<br />

Ease ≥ 2 55% (85/155) 46% (424/912) 1.5 : 1.1-2.2 P < 0.05<br />

Ease ≥ 3 27% (42/155) 23% (208/912) 1.3 : 0.9-2.0 P = 0.17<br />

Ease ≥ 4 11% (17/155) 9% (82/912) 1.4 : 0.8–2.5 P = 0.32<br />

Ease ≥ 5 5% (7/155) 4% (32/912) 1.4 : 0.6–3.4 P = 0.44<br />

190


Table 7.10: The number <strong>of</strong> cows with and without bovine respiratory disease in the 60 days<br />

following enrollment (BRD60) from each source farm in the analysis <strong>of</strong> first milk test, the leastsquares<br />

mean (± standard error <strong>of</strong> the mean) age at first milk test and the effect <strong>of</strong> BRD60 on<br />

first milk test date.<br />

Source<br />

Farm<br />

BRD60 +<br />

(n=)<br />

BRD60 –<br />

(n=)<br />

Average<br />

days in<br />

milk at first<br />

DHIA test<br />

Effect <strong>of</strong> BRD60 on Milk<br />

Test Weight (LSMean ±<br />

S.E.M)<br />

BRD- BRD+<br />

191<br />

P-value<br />

1 36 232 33 ± 17 28.9 ± 0.5 26.0 ± 1.1 0.01<br />

2 19 74 17± 12 31.6 ± 1.1 29.1 ± 1.5 < 0.001<br />

3 18 129 21 ± 10 30.3 ± 0.6 28.5 ± 1.6 0.29<br />

4 12 85 25 ± 13 26.5 ± 0.7 28.1 ± 1.9 0.42<br />

5 54 290 18 ± 11 26.9 ± 0.4 26.3 ± 0.9 0.48<br />

6 14 86 19 ± 12 27.4 ± 0.7 31.1 ± 1.8 0.05


Table 7.11: Distribution <strong>of</strong> heifers that survived to 120 days post-calving recorded by calving<br />

ease score and the probability <strong>of</strong> survival relative to a calving ease score <strong>of</strong> 1<br />

Calving Ease<br />

Score<br />

Proportion <strong>of</strong> heifers that<br />

survived to 120 days in milk<br />

Odds Ratio: 95%<br />

Confidence Interval<br />

relative to score = 1<br />

192<br />

P-value<br />

1 95.4% (519/544)<br />

2 93.2% (234/251) 0.5: 0.3-1.0 0.05<br />

3 92.5% (137/148) 0.5 : 0.2-1.0 0.05<br />

4 91.7% (55/60) 0.3 : 0.1-0.9 0.03<br />

5 92.3% (36/39) 0.2 : 0.1-0.9 0.04


Figure 7.1: Timeline <strong>of</strong> trial events<br />

193


Figure 7.2: Difference in least squares means (± SE ) <strong>of</strong> the weight (kg) <strong>of</strong> heifers with BRD60 1<br />

relative to calves without BRD60 at 5 time points at a commercial heifer raising facility,<br />

controlling for birth weight, age at measurement and season with random effects for source farm<br />

and weekly enrolment cohort.<br />

Difference in Body weight (KG)<br />

5<br />

0<br />

-5<br />

-10<br />

-15<br />

-20<br />

-25<br />

56 98 181 269 382<br />

** P < 0.001<br />

Weaning<br />

Barn<br />

Transition<br />

Barn<br />

Rearing Barn<br />

**<br />

Grower<br />

Barn 1<br />

1 BRD60: Bovine respiratory disease occurring within 60 days <strong>of</strong> study enrolment<br />

**<br />

194<br />

Grower<br />

Barn 2<br />

**<br />

Average age (days) at measurement<br />

Breeding<br />

Barn<br />

**


Figure 7.3: Differences <strong>of</strong> least squares means (± SE ) <strong>of</strong> the weight (kg) <strong>of</strong> heifers with no prior<br />

history <strong>of</strong> bovine respiratory disease that received tulathromycin at study enrolment relative to<br />

heifers that received oxytetracycline at study enrolment at entry at 5 time points at a commercial<br />

heifer raising facility. The linear mixed models controlled for birth weight, age at measurement<br />

and season as fixed effects and for source farm and weekly enrolment cohort as random effects.<br />

Difference in Body weight (KG)<br />

10<br />

9<br />

8<br />

7<br />

6<br />

5<br />

4<br />

3<br />

2<br />

1<br />

0<br />

-1<br />

Weaning Barn Transition Barn Grower<br />

Barn 1<br />

Rearing Barn<br />

** P < 0.001<br />

56 98 181 269 382<br />

Average age (days) at measurement<br />

**<br />

**<br />

195<br />

**<br />

Grower<br />

Barn 2<br />

**<br />

Breeding<br />

Barn


Figure 7.4: Time to event analysis for age at first calving stratified by BRD60 status with culled<br />

heifers censored (P = 0.02).<br />

196


Appendix 7.1: Daily minimum and maximum temperature at the Rochester, New York weather<br />

station, located approximately 65 km from study site, by season.<br />

40<br />

30<br />

20<br />

10<br />

0<br />

-10<br />

-20<br />

Reference<br />

Season Season 2 Season<br />

3<br />

Bach, A. 2011. Associations between several aspects <strong>of</strong> heifer development and dairy cow<br />

survivability to second lactation. J. Dairy Sci. 94:1052-1057.<br />

197<br />

Max TemperatureC<br />

Min TemperatureC


CHAPTER 8: CONCLUSIONS AND FUTURE RESEARCH<br />

8.1 GENERAL DISCUSSION<br />

There are many challenges associated with rearing healthy dairy heifer calves that can<br />

result in high mortality rates and welfare problems. Multidisciplinary approaches <strong>of</strong> evaluating<br />

management practices, identifying animals at risk <strong>of</strong> disease, and quantifying the impact <strong>of</strong><br />

diseases are valuable tools for improving calf welfare and productivity. High mortality rates in<br />

dairy calves are important concerns for producers, due to the monetary costs <strong>of</strong> calf loss, the<br />

welfare costs to animals during illness and convalescence, and the impact <strong>of</strong> mortality and calf<br />

suffering on public perception <strong>of</strong> the dairy industry. Two diseases are primarily responsible for<br />

mortality in neonatal dairy calves: neonatal calf diarrhea complex (diarrhea) and bovine<br />

respiratory disease complex (BRD). These diseases are responsible for 79% <strong>of</strong> the deaths in milk<br />

fed calves and 59% <strong>of</strong> the deaths in weaned dairy calves (USDA, 2007). To reduce the impact <strong>of</strong><br />

these two diseases, new and improved methods <strong>of</strong> disease identification and animal management<br />

are needed. The three main areas <strong>of</strong> focus <strong>of</strong> this doctoral thesis are the use <strong>of</strong> behavior to<br />

identify animals at risk <strong>of</strong> disease, group level management <strong>of</strong> disease risk, and identification <strong>of</strong><br />

long-term impacts <strong>of</strong> BRD.<br />

Animal diseases have many negative impacts on animal welfare and production (Chapter<br />

1). The early identification <strong>of</strong> animals at risk <strong>of</strong> disease allows for interventions either through<br />

reducing or eliminating risk factors. Examples include altering weaning management practices<br />

(Chapter 4), or in the absence <strong>of</strong> controllable risks, such as poor weather, administering a<br />

prophylactic antibiotic (Chapter 5 and 6). Even if risk factors are reduced, some disease still<br />

occurs. Therefore, to improve recovery and reduce welfare impacts, early identification <strong>of</strong><br />

198


clinical diseases is important. For this reason, behavior alterations associated with sickness are <strong>of</strong><br />

value to identify at risk animals. This thesis identified two predictors <strong>of</strong> poor growth in<br />

individually reared calves; 1. altered postures in calves under 2 weeks <strong>of</strong> age, 2. high lethargy<br />

scores during clinical diarrhea in calves under 2 weeks and calves between 4 and 6 weeks <strong>of</strong> age<br />

(Chapter 2). These measures <strong>of</strong> behavior were originally selected to identify calves with low<br />

motivation to explore their environment and decreased appetite based on known motivational<br />

changes associated with sickness (Aubert, 1999) as well as indicators <strong>of</strong> visceral pain, previously<br />

identified in lambs (Molony et al., 2002). Poor growth was evaluated based on its association<br />

with disease in neonatal dairy calves (Chapter 5). The association <strong>of</strong> posture and lethargy scores<br />

with poor growth supports the hypothesis that these are indicators <strong>of</strong> sickness in individually<br />

housed dairy calves. The main challenge for this study was that behavioral measures were taken<br />

every other week, which may have allowed for illness and recovery between measures,<br />

decreasing the sensitivity <strong>of</strong> these tests. In the future, this could be addressed by measuring the<br />

selected behaviors in a smaller, more intensive study in which daily behavioral scores and<br />

weekly body weights are measured. However, the behavior measures examined here do provide<br />

a valuable starting point for the identification <strong>of</strong> sickness behaviors that are most beneficial for<br />

identifying disease. These behavior measures should be further refined for use by calf caretakers<br />

to improve disease recognition.<br />

Milk fed calves are at greatest risk <strong>of</strong> diarrhea (USDA, 2010). However, a second disease<br />

risk period, frequently associated with BRD, is weaning and the change to group housing. For<br />

this thesis, poor post-weaning growth was chosen as an indicator <strong>of</strong> poor adaptation to weaning<br />

and movement based on the hypothesis that these animals are the most likely to contract BRD.<br />

199


Heifers with poor post-weaning growth were identified as those that were most active in the<br />

three days following weaning (Chapter 3) and that spent greater amounts <strong>of</strong> time at the bunk<br />

without eating (bunk engaged behavior). The effect <strong>of</strong> weight gain and lying distance was<br />

confounded by nursery housing system and location; Calves from Nursery 2, who had been<br />

housed in individual outdoor calf hutches, gained less weight and spent more time lying far from<br />

other calves compared to calves from Nursery 1, who had been housed in adjacent individual<br />

pens in a barn. This resulted in limitations to our ability to interpret the results, but does suggest<br />

that social lying distance is associated with weight gain and nursery housing system may impact<br />

weight gain and behaviour. From these findings several areas <strong>of</strong> study were identified. The<br />

variation in behavioral response observed in this study suggests that there is potential to<br />

intervene prior to weaning to improve adaptation and reduce stress associated with these<br />

experiences. For this reason, focusing future research on preparing calves for weaning by<br />

decreasing stress and encouraging calf starter intake are important areas <strong>of</strong> focus for disease risk<br />

management and animal welfare.<br />

Disease management is an important area <strong>of</strong> study for the welfare and productivity <strong>of</strong><br />

dairy calves. There are many approaches to reducing disease risk; in this thesis management<br />

systems currently in use in modern dairy farms were evaluated. Two approaches were used to<br />

reduce the risk <strong>of</strong> calves succumbing to clinical disease, the first was reduction <strong>of</strong> disease<br />

susceptibility through the use <strong>of</strong> graduated weaning (Chapter 4) and the second was the use <strong>of</strong><br />

prophylactic antibiotics at a group level during periods <strong>of</strong> high risk (Chapter 5 and 6). Graduated<br />

weaning programs have been promoted to reduce the stress <strong>of</strong> dairy calves during the post-<br />

weaning and movement time period. The findings from this experiment were that mixing calves,<br />

200


and movement to a novel environment was associated with increased steps and decreased lying<br />

time. If mixing and movement to a novel environment occurred a week apart the increase in the<br />

number <strong>of</strong> steps and the decrease in lying time was reduced compared to mixing and movement<br />

occurring simultaneously. Further research in needed to determine if the decreased activity in the<br />

novel environment performed by pre-mixed calves was a result <strong>of</strong> decreases agonistic behaviours<br />

or if the agonistic behaviours occurred whenever calves were mixed. This information would<br />

determine if premixing calves reduces the stress associated with movement to a novel<br />

environment. Further study <strong>of</strong> this management procedure would be valuable to identify the<br />

types <strong>of</strong> behaviors occurring upon movement to a novel environment and determine the effect <strong>of</strong><br />

increased transportation stress and grouping on adaptation to a novel environment and the risk <strong>of</strong><br />

disease.<br />

A second approach, which can be used in conjunction with minimizing weaning stress, is<br />

the strategic use <strong>of</strong> antibiotics at the group level. The primary criteria for consideration <strong>of</strong> this<br />

approach were the use <strong>of</strong> antibiotics during a short time period <strong>of</strong> high risk <strong>of</strong> disease that cannot<br />

be eliminated through other methods, and the welfare impacts <strong>of</strong> disease, particularly when<br />

recovery is poor or prolonged. Prophylactic antimicrobial use was reviewed based on its<br />

established efficacy in beef feedlots and its previously unstudied use in dairy calves. Two time<br />

periods were evaluated; the first was shortly after birth when calves arrived at a heifer raising<br />

facility and the second was upon first movement to group housing following weaning.<br />

Prophylactic antimicrobial use reduced disease in both populations. For dairy calves that<br />

received a long acting antibiotic upon arrival at a heifer raising facility, the incidence <strong>of</strong> otitis<br />

and diarrhea were reduced in the first 8 weeks <strong>of</strong> life. For this reason, the efficacy criterion for<br />

201


strategic antimicrobial use was met. However, otitis, the targeted disease, had minimal effect on<br />

growth, indicating that on this farm, otitis did not have substantial long-term impacts. In<br />

addition, otitis could not be definitively linked with M. bovis, an organism previously associated<br />

with severe BRD, arthritis and otitis media. Furthermore, the incidence <strong>of</strong> M. bovis was not<br />

different between antibiotic and placebo treatment groups. Although culture methods used in this<br />

study limited interpretations, the minimal impact <strong>of</strong> both diarrhea and otitis on growth suggest<br />

that this disease does not meet our stated criteria for responsible prophylactic antimicrobial use<br />

<strong>of</strong> a severe disease with poor recovery rates. In addition, the small difference in ADG (0.02 kg)<br />

between calves that received the antibiotic compared to the control group also indicates a small<br />

impact. When prophylactic antimicrobial use was evaluated in 8 week old calves after movement<br />

to group housing, the incidence <strong>of</strong> BRD was reduced and calves that received TUL had 0.12 kg<br />

higher ADG compared to control animals, provided there was no prior history <strong>of</strong> BRD (Chapter<br />

6). The lack <strong>of</strong> growth effect in calves with a prior history <strong>of</strong> BRD provided important<br />

information for the efficacy <strong>of</strong> this approach and indicated that the benefits <strong>of</strong> prophylactic<br />

antimicrobial use may be reduced in calves with a history <strong>of</strong> BRD. An additional benefit was that<br />

the strategic use <strong>of</strong> a long-acting antibiotic improved the probability <strong>of</strong> surviving to first lactation<br />

for calves treated for BRD after movement to group housing.<br />

Long-term effects <strong>of</strong> disease are important to evaluate not just for productivity, but to<br />

identify long-term implications which may compromise animal welfare. The <strong>final</strong> objective <strong>of</strong><br />

this thesis program was to update and identify the long-term implications <strong>of</strong> this disease (Chapter<br />

7). BRD was found to decrease ADG for almost 6 months after the initial disease period,<br />

resulting in a 15 kg difference in body weight between affected and unaffected animals at 9<br />

202


months <strong>of</strong> age. The lower growth rate indicates that calves recover very slowly from BRD, and<br />

that BRD has potentially altered the calves’ physiology or behavior either through decreased<br />

feed efficiency or decreased feed intake. The long-term effects <strong>of</strong> BRD were further<br />

demonstrated by the lower probability that calves with BRD would survive to first calving, the<br />

greater age at first calving and the higher risk <strong>of</strong> some level <strong>of</strong> dystocia at first calving, relative<br />

to calves that did not experience BRD during the 60 days following movement to group housing.<br />

The effect <strong>of</strong> BRD on milk production was less clear due to a likely healthy worker bias resulting<br />

from the high rate <strong>of</strong> removal <strong>of</strong> BRD heifers prior to calving. To avoid this bias, a different<br />

approach may be required in that the animal’s genetic potential for milk production should be<br />

accounted for in the model to determine if BRD decreases milk production. This evaluation<br />

would require a larger sample size than was available in this study since it would require many<br />

daughters sharing the same sire.<br />

The results <strong>of</strong> these three thesis objectives have demonstrated that a multi-disciplinary<br />

approach to improving management <strong>of</strong> dairy calf health and the identification <strong>of</strong> long-term<br />

impacts <strong>of</strong> disease are beneficial. Future research should continue to address the issues <strong>of</strong> early<br />

disease identification, risk management and animal health in order to improve the productivity<br />

and welfare <strong>of</strong> dairy calves.<br />

8.2 FUTURE RESEARCH<br />

Although several management procedures that are associated with reducing the risk <strong>of</strong><br />

dairy calf disease, and some behaviors associated with poor growth were investigated in this<br />

thesis, these findings also illustrated many other areas <strong>of</strong> investigation that need to be considered.<br />

203


Specifically,<br />

How can disease severity be appropriately identified to improve animal care?<br />

Bovine respiratory disease complex and neonatal calf diarrhea have been associated with<br />

many long-term impacts on calf health and welfare which brings into question the recovery rate<br />

<strong>of</strong> these diseases and raises the potential for altered management to improve recovery. However,<br />

for this to be best utilized, the calves that are most severely affected need to be identified.<br />

Currently there is a scoring system reported in literature which attempts to categorize animals<br />

based on clinical signs (Lago et al., 2006). However, this scoring system has not been formally<br />

validated. The signs selected include nasal and ocular discharge, rectal temperature, and ear<br />

droop. Behavioral indicators <strong>of</strong> illness are not included in this score. Defined behavioral<br />

indicators may be useful for identifying animals that are in the greatest discomfort and most<br />

severely affected, which is not provided in the current scoring system. In order to validate an<br />

effective scoring system and to determine cut points for mortality risk, a cohort study in which<br />

calves are scored during clinical disease and the number <strong>of</strong> re-treatments are recorded should be<br />

conducted. Animals would then be monitored until calving, in order to determine which cut-<br />

points and signs are the most sensitive and specific for identifying severely affected animals.<br />

From this information, recovery information for different cut-points can be collected and<br />

improved treatment and management decisions may be made.<br />

Why is the growth <strong>of</strong> dairy calves affected for a prolonged time period?<br />

There are two potential mechanisms for the poor growth <strong>of</strong> heifers with BRD: decreased<br />

feed intake and feed conversion. These two factors may also be affected by social factors<br />

including the ability <strong>of</strong> heifers affected by BRD to compete at the feed bunk. In order to improve<br />

204


the management <strong>of</strong> heifers recovering from BRD, there is a need to determine which factors<br />

impair their growth. There are four hypotheses for this decreased weight gain observed in heifers<br />

with BRD after recovery from clinical disease;<br />

1) decreased feed efficiency,<br />

2) decreased feed intake,<br />

3) fewer, shorter feeding bouts,<br />

4) more displacements at the feed bunk.<br />

Each <strong>of</strong> these hypotheses should be tested at different stocking densities and social<br />

mixing to determine if BRD heifers at high stocking densities show greater reductions in feed<br />

efficiency and intake, greater alterations in feeding behavior, and more feed bunk displacements<br />

relative to healthy heifers and to BRD heifers housed at lower stocker density.<br />

If the outcomes <strong>of</strong> feeding behavior or feed efficiency are altered by stocking density,<br />

there is a potential to alter the management <strong>of</strong> dairy calves during convalesce based on this<br />

knowledge. The impact <strong>of</strong> different proportions <strong>of</strong> calves with a prior history <strong>of</strong> BRD within a<br />

group on disease spread, relapse rates and growth was recently investigated at a commercial<br />

heifer rearing facility in Spain (Bach et al., 2011). These researchers did not find a difference in<br />

ADG depending on group composition. However, their analysis was completed at the pen level,<br />

and using a small sample size. Further research to determine the impact <strong>of</strong> group composition on<br />

the growth <strong>of</strong> sick calves would be beneficial to identify improved methods <strong>of</strong> managing animals<br />

during convalescence.<br />

205


Does transportation, initial social mixing and a novel environment in conjunction with<br />

weaning result in additive stressors, and how far apart in time must these stressors occur to<br />

reduce stress and avoid the additive effect?<br />

This thesis began to investigate the impact <strong>of</strong> a graduated weaning practice on the<br />

behavior <strong>of</strong> dairy calves. However, this was a pilot study and many more variables need to be<br />

understood before this system can be recommended or discouraged on commercial farms. Three<br />

main questions must be addressed for this program to be understood; 1) which stressors result in<br />

additive stress?, 2) what is the duration <strong>of</strong> response to the individual stressors?, 3) what is<br />

minimum length <strong>of</strong> time required between each <strong>of</strong> the stressors to avoid inflicting additive<br />

stress? These questions would best be answered using controlled experiments that incorporate<br />

behavior, together with physiological responses, specifically heart rate, cortisol levels with an<br />

ACTH challenge to detect stress induced immunosuppression, and weight gain. The stressors<br />

investigated should include transportation stress in weaned calves, first social mixing, removal <strong>of</strong><br />

milk from the diet, dehorning, and change in environment. Although removal <strong>of</strong> milk from diet<br />

and dehorning have been extensively studied in dairy calves, the effect <strong>of</strong> change in<br />

environments, transportation, and first social mixing need further investigation to determine how<br />

they interact with the above stated stressors. In addition, there is some suggestion from the<br />

results <strong>of</strong> Chapter 3 that housing may play a role in adaption to a novel environment. As such,<br />

the stress responses <strong>of</strong> calves reared in hutches versus nursery pens should be compared when<br />

subjected to the above stressors.<br />

206


8.3 REFERENCES<br />

Aubert, A. 1999. Sickness and behaviour in animals: A motivational perspective. Neuroscience<br />

& Biobehavioral Reviews. 23:1029-1036.<br />

Bach, A., C. Tejero, & J. Ahedo. 2011. Effects <strong>of</strong> group composition on the incidence <strong>of</strong><br />

respiratory afflictions in group-housed calves after weaning. J. Dairy Sci. 94(4): 2001-2006.<br />

Lago, A., S. M. McGuirk, T. B. Bennett, N. B. Cook and K. V. Nordlund. 2006. Calf respiratory<br />

disease and pen microenvironments in naturally ventilated calf barns in winter. J. Dairy Sci.<br />

89:4014-4025.<br />

Molony, V., J. E. Kent and I. J. McKendrick. 2002. Validation <strong>of</strong> a method for assessment <strong>of</strong> an<br />

acute pain in lambs. Appl. Anim. Behav. Sci. 76:215-238.<br />

USDA. 2010. Dairy 2007, Heifer Calf Health and Management Practices on U.S. Dairy<br />

Operations, 2007. USDA:APHIS:VS, CEAH., Fort Collins, CO. #550.0110.<br />

USDA. 2007. Dairy 2007, Part I: Reference <strong>of</strong> Dairy Cattle Health and Management Practices in<br />

the United States. 2007 USDA-APHIS-VS, CEAH., Fort Collins, CO. #N480.1007.<br />

207

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!